effectiveness of streambank-stabilization techniques along ...effectiveness of...

Prepared in cooperation with the Alaska Department of Environmental Conservation and Alaska Department of Fish and Game

Effectiveness of Streambank-Stabilization Techniques Along the Kenai River, AlaskaWater-Resources Investigations Report 99-4156

U.S. Department of the InteriorU.S. Geological Survey

Cover: View of site near river mile 21, Kenai River, Alaska

Effectiveness of Streambank-Stabilization Techniques

Along the Kenai River, Alaska

By Joseph M. Dorava__________________________________________________________________

U.S. GEOLOGICAL SURVEY

Water-Resources Investigations Report 99-4156

Prepared in cooperation with the

ALASKA DEPARTMENT OF ENVIRONMENTAL CONSERVATION

and

ALASKA DEPARTMENT OF FISH AND GAME

Anchorage, Alaska 1999

U.S. DEPARTMENT OF THE INTERIORBRUCE BABBITT, Secretary

U.S. GEOLOGICAL SURVEYCharles G. Groat, Director

Use of trade names in this report is for identification purposes only and does not constitute endorsementby the U.S. Geological Survey.

For additional information: Copies of this report may be purchased from:

District Chief U.S. Geological SurveyU.S. Geological Survey Branch of Information Services4230 University Drive, Suite 201 Box 25286Anchorage, AK 99508-4664 Denver, CO 80225-0286

http://ak.water.usgs.gov

CONTENTS

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Objectives and Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Previous Boatwake Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Streambank-Stabilization Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Site Selection and Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

River Mile 16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

River Mile 21. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Wake Heights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Bank Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Habitat Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Water Velocity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

Protective Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

References Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

FIGURES

1. Map showing location of Kenai River, Alaska, and sampling sites at river miles 16 and 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2. Photograph showing aerial view of river mile 16, Kenai River, Alaska . . . . . . . . . . 6

3. Photograph showing bio-log installed near river mile 16, Kenai River, Alaska . . . . 7

4. Photograph showing spruce trees cabled to streambank near river mile 16,Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

5. Photograph showing aerial view of river mile 21, Kenai River, Alaska . . . . . . . . . . 9

6. Photograph showing view of site near river mile 21, Kenai River, Alaska:(A) natural streambank and (B) engineered bank stabilization technique . . . . . . . . . 11

7. Graph showing height of wakes striking the bank near river mile 16 of the Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

8. Graph showing weight of material eroded from the bank near river mile 16 of the Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

9. Graph showing weight of material eroded from the bank near river mile 21 of the Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

10. Photograph showing Christmas trees being used as shoreline erosioncontrol in Louisiana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

TABLES

1. Wake heights measured near river mile 16, Kenai River, Alaska . . . . . . . . . . . . . . . .102. Weight of eroded bank material retained in collection pan near river miles

16 and 21, Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123. Weight of eroded bank material retained in collection pan during, before, and

after boatwake tests near river miles 16 and 21, Kenai River, Alaska. . . . . . . . . . . . .144. Amount of protective cover for juvenile salmon at river miles 16 and 21,

Kenai River, Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

CONVERSION FACTORS

Multiply by To obtain

inch 25.4 millimeter

foot 0.3048 meter

mile 1.609 kilometer

square foot 0.09290 square meter

cubic foot 0.02832 cubic meter

foot per second 0.3048 meter per second

mile per hour 1.609 kilometer per hour

cubic foot per second 0.02832 cubic meter per second

pound per foot 1.488 meter per foot

Effectiveness of Streambank-Stabilization Techniques Along the Kenai River, Alaska

By Joseph M. Dorava

ABSTRACT

The Kenai River in southcentral Alaska isthe State’s most popular sport fishery and aneconomically important salmon river that gen-erates as much as $70 million annually. Boat-wake-induced streambank erosion and theassociated damage to riparian and riverine hab-itat present a potential threat to this fishery.Bank-stabilization techniques commonly in usealong the Kenai River were selected for evalua-tion of their effectiveness at attenuating boat-wakes and retarding streambank erosion.Spruce trees cabled to the bank and biodegrad-able man-made logs (called “bio-logs”) pinnedto the bank were tested because they are com-monly used techniques along the river. Thesetwo techniques were compared for their abilityto reduce wake heights that strike the bank andto reduce erosion of bank material, as well as forthe amount and quality of habitat they providefor juvenile chinook salmon. Additionally, anengineered bank-stabilization project was eval-uated because this method of bank protection isbeing encouraged by managers of the river. Dur-ing a test that included 20 controlled boatpasses, the spruce trees and the bio-log provideda similar reduction in boatwake height and bankerosion; however, the spruce trees provided agreater amount of protective habitat than thebio-log. The engineered bank-stabilizationproject eroded less during nine boat passes andprovided more protective cover than the adja-cent unprotected natural bank. Features of thebank-stabilization techniques, such as tree limbsand willow plantings that extended into thewater from the bank, attenuated the boatwakes,which helped reduce erosion. These featuresalso provided protective cover to juvenilesalmon.

INTRODUCTION

Background

Streambank erosion and sedimentationhave been recognized as substantial non-pointsources of pollution throughout the nation(Duijsings, 1986; Lietman and others, 1984;Odgaard, 1984; U.S. Army Corps of Engineers,1981; von Guerard, 1989). Furthermore, thesegeomorphic processes are contributing causesfor significant salmon population declines inanadromous river systems (American Societyof Civil Engineers, 1992; Beschta, 1989; Bjorn,1969; Hansen, 1971; Klingeman and others,1990; Meehan, 1974; Meehan and Swanston,1977). Riparian areas that provide cover andfood sources for juvenile salmon are destroyedby streambank erosion, which subsequentlyincreases the transport and deposition of sedi-ment. Spawning gravels thereby becomeclogged with deposits of fine sediment. Theeffect is a reduction of salmon numbersbecause of habitat loss (Beschta, 1989). Inaddition, many bank-protection, stabilization,or restoration techniques currently in use pro-vide poor fish habitat. For example, concreteblock retaining walls can very effectively pro-tect a bank from eroding; however, the smoothflat face of the wall provides little habitat suit-able for fish.

Boatwakes cause bank erosion and resus-pend streambed sediments, which can degradewater quality (Byrd and Perona, 1980; Dez-man, 1990; Dorava and Moore, 1997; Garradand Hey, 1987 and 1988; Jackivicz andKuzminski, 1973a, 1973b; Nanson and others,1994; Schoellhamer, 1990; Smart and others,

Introduction 1

oret-

tatep-

dnk-edr,

to

i-teiletick-gh- indr-s

ti-dalo-d

ye-heenk

uit-

jec-nts:

esat-

1985; von Krusenstierna, 1990; Yousef andothers, 1978). Boatwake-induced erosion andhabitat loss in important salmon-rearing areascan reduce the number of salmon ultimatelyavailable to fishermen and threaten the fishery.Bank-stabilization techniques and erosion con-trol measures are commonly used to protectwater quality and fisheries, but have had variedsuccess (Dezman, 1990; Garcia, 1988; Nel-son,1986; Trimble, 1997).

The fisheries in the Kenai River in south-central Alaska (fig. 1) generate as much as $70million annually (Liepitz, 1994). Because ofthe economic importance of these fisheries,managers of the river are concerned about any-thing that may threaten the fishery. Paramountamong their concerns are those activities suchas accelerated bank erosion from boatwakesand streamside development that are directlyrelated to fish and fish habitat. Chinook salmonare the primary target species for sport fisher-men on the Kenai River. Rearing habitat forjuvenile chinook salmon is one of the principalfactors limiting production of this fishery. Juve-nile chinook salmon in the Kenai River com-monly rear in areas within about 6 feet of thestreambank, where water velocities are lessthan 1 foot per second, stream substrate isgreater than 1.5 inches in diameter, and abun-dant cover can protect them from predators(Bendock and Bingham, 1988; Burger and oth-ers, 1982; Estes and Kuntz, 1986; Liepitz,1994; Raleigh and others, 1986). These rearinghabitat features are easily altered by boatwake-induced streambank erosion and streamsidedevelopment. In particular, the cover providedby streamside vegetation is commonlyremoved by development or destroyed by stre-ambank erosion.

The Kenai River salmon fisheries may bethreatened by streambank erosion generated byboatwakes (Dorava and Moore, 1997; Recken-dorf, 1991; Scott, 1982). As the State’s popula-tion and river use increase, boatwake-inducedstreambank erosion and the associated riparian

and riverine habitat losses most likely will alsincrease. Bank-stabilization techniques that aeffective at reducing bank erosion by attenuaing the effects of boatwakes and provide habifor rearing juvenile chinook salmon would bvaluable to the economy, to streamside proerty owners, and to wildlife, river, and lanmanagement agencies. Numerous streambastabilization techniques have been developand deployed along the Kenai River; howevefew of these have been designed specificallyprovide protection from boatwakes.

The ultimate design goal of bank stabilzation is not to minimize erosion, but to creaa maintainable environment, because mobnatural stream channels are dynamic, not sta(Thomas, 1990). During this study, three banstabilization techniques commonly used alonthe Kenai River were evaluated. These tecniques were compared for their effectivenessproviding protection from boatwake-inducedamage to riparian and riverine habitat impotant to juvenile chinook salmon. The study wasupported in part by funds designated for migation of non-point source pollution provideby the Alaska Department of EnvironmentConservation, and for habitat restoration prvided by the Alaska Department of Fish anGame.

Objectives and Approach

The two primary objectives of the studwere (1) to evaluate how effectively three strambank-stabilization techniques reduce tmaximum height of boatwakes striking thbank and retard boatwake-induced streambaerosion and (2) to evaluate the amount and sability of potential fish habitat provided bythose techniques.

The approach used to address these obtives was to make three types of measureme

(1) The average maximum height of boatwakstriking the bank was measured. The bo

2 Effectiveness of Streambank-Stabilization Techniques Along the Kenai River, Alaska

Introduction 3

ub-

n-d-ticr-

s,l

dof),a

ndithat-asn-ger-rse.

ri-g

60tateditumaterat-

tra-hofionedheer-lyent

wakes were measured with a float encasedin a screen mesh and attached to a paperchart recorder. These wake-recordinginstruments were also used in previousstudies to evaluate the effects of boatwakeson streambank erosion and are described ingreater detail by Dorava and Moore(1997).

(2) The average weight of material eroded fromeach foot of the streambank was measured.The material was collected in a 12- by 18-by 2-inch baking pan pinned to the streambottom near the base of the bank with the12-inch side parallel with the streambank.The average weight of material collectedin the pan while no boatwakes were strik-ing the bank was subtracted from theweight of material collected in the pan dur-ing each boat pass. The amount of materialcollected in the pan may have been lessthan was eroded from the bank. Somematerial may not have stayed in the panbecause of the washing action of numerouswakes that occur with a single boat pass.This pan-collection method was also usedby Dorava and Moore (1997) to evaluatethe effects of boatwakes on streambankerosion.

(3) The amount of juvenile chinook salmonhabitat available for protective cover wasmeasured by determining the area of pro-tective cover provided by each stabiliza-tion technique.

Previous Boatwake Studies

Most research that specifically addressesboatwake-induced streambank erosion is doneon large rivers such as the Mississippi, Illinois,or Ohio Rivers, and generally only for largecommercial vessels such as navy ships or cargobarges (Bhowmik and Demissie, 1982; Das andJohnson, 1970; Hagerty and others, 1981; Kar-aki and vanHoften, 1974; Sparks, 1975). Espe-cially rare are studies of smaller rivers such as

the Kenai River, whose primary traffic is 10- to26-foot-long recreational boats (Dorava andMoore, 1997). Studies on rivers in England—where recreational and pleasure boats are a sstantial component of overall boat traffic—documented that turbidity and sediment cocentration increased, the river channel wiened, and densities of submerged macrophyplants declined as a result of boat traffic (Garad and Hey, 1987, 1988; Hilton and Phillip1982; Murphy and Eaton, 1983). Additionastudies of recreational boat traffic in OlTampa Bay and the intra-coastal waterway Florida (Schoellhamer, 1990; Verdon, 1999and in Chesapeake Bay in Maryland (Zabawand Ostrom, 1980) also report shoreline abed-sediment disturbances associated wboatwakes. The Minnesota Department of Nural Resources (Bentley and others, 1991) hpublished a brochure containing design, costruction, and permitting details for protectinstreambanks from erosion, but noted that pformance data for these techniques were spa

During investigations of the River Burein England, Garrad and Hey (1987) tested vaous size boats (12 to 42 feet long) travelinalong various sailing lines (about 0, 30, and feet from the bank) and moving at differenspeeds (3 to 7 miles per hour). They found thsuspended-sediment concentration increasrapidly following passage of a boat when traveled at a speed greater than some minimthreshold speed. Furthermore, at speeds grethan this threshold speed, the increase in bowake-induced suspended-sediment concention dissipated slowly. Because of the higfrequency of boat passes, the effects increased suspended-sediment concentratfrom a single boat pass were not dissipatbefore additional boat passes occurred. Tresults imply that even when the boats are opated within prescribed speed limits, potentialdetrimental increases in suspended-sedimconcentrations can occur.


There may be a direct application of thisEnglish study to the Kenai River. Although theKenai River does not currently have boat-oper-ating speed limits, the existing 35-horsepowerlimit on outboard motors can be considered, atleast conceptually, in a similar manner. It illus-trates that significant streambank erosion andsubsequent habitat losses can occur even whenboats are operated within the current regulatoryframework.

STREAMBANK-STABILIZATION TECHNIQUES

Three popular streambank-stabilizationmaterials and techniques in use along the KenaiRiver were selected for this study. Theseincluded (1) man-made bio-degradable logsmade of coconut husks and woven into a roundcylinder (bio-logs); (2) freshly cut spruce treescabled together in a bunch; and (3) an engi-neered combination of bio-logs and cabledspruce trees installed together with willowplantings. A database of restoration projectsmaintained by the Kenai River Center in Sol-dotna indicates that in 1998, about 6,615 feet ofthe Kenai River streambank was stabilized withspruce trees cabled to the bank, about 6,730 feetof the bank was stabilized with bio-logs, andabout 1,700 feet of the bank was stabilized withan engineered combination of materials includ-ing spruce trees, bio-logs, and willow plantings(Dean Hughes, Alaska Department of Fish andGame, written commun., 1998). The threebank-stabilization materials and techniqueswere evaluated for their effectiveness at attenu-ating boatwakes and protecting the bank fromerosion, and compared for the amount and qual-ity of habitat they provided to juvenile chinooksalmon, primarily as protective cover. At onesite along the river, both bio-logs and sprucetrees were installed in front of the bank and aboatwake-recording instrument. These tech-niques were compared for their ability toreduce the maximum height of boatwakes strik-

ing the bank and for their ability to retard stre-ambank erosion. At another site about 5 milesupstream, the engineered combination of bio-logs, cabled spruce trees, and willow plantingswas evaluated primarily for its ability to retardstreambank erosion and provide protectivecover to juvenile chinook salmon. It wasimpossible to quantify the ability of this engi-neered site to attenuate boatwakes because itsprevious installation into the streambank pre-vented measurement of wake characteristics.

SITE SELECTION AND DESCRIPTION

River Mile 16

A site in the lower river near river mile16, downstream from the city of Soldotna, wasselected for study (figs. 1 and 2). Boatwake anderosion recording instruments have beendeployed at this site since 1995, and substantialboatwake-induced streambank erosion hasoccurred here (Dorava and Moore, 1997). Thissite is on the inside of a meander bend and isheavily vegetated with alder, willow, andmature spruce trees. The low banks at thisinstrumented site provided an opportunity totest bank response to boatwakes under the fol-lowing conditions: (1) no protection, (2) a tem-porarily installed bio-log, and (3) temporarilyinstalled spruce trees.

A bio-log and cabled spruce trees wereinstalled near river mile 16 on August 4, 1998.During the wake testing, stream discharge atthe site was about 13,800 cubic feet per secondbased on records at the nearby USGS stream-gaging station Kenai River at Soldotna (No.15266300; fig. 1) (Bertrand and others, 1999).This discharge is within 5 percent of the long-term mean discharge for August (14,470 cubicfeet per second), which has typically been themonth of maximum mean discharge. This 5percent discharge difference translates into awater depth difference of about 0.15 foot, indi-cating that the river flow during the August test

Streambank-Stabilization Techniques 5

period was close to mean annual high-waterconditions. The bio-log, about 20 feet long and12 inches in diameter, was installed in front ofthe bank by pinning it to the stream bottom withmetal stakes (fig. 3). The bio-log protrudedabout 3 inches above the waterline. Two freshly

cut spruce trees about 20 feet long were cabledtogether and attached to the streambank withshort cables (fig. 4). The spruce trees extendedabove the water line and into the river consid-erably more than the bio-log.


Site Selection and Description 7

River Mile 21

Another site near river mile 21, upstreamfrom the Sterling Highway Bridge in Soldotna,was also selected for study (figs. 1 and 5). Boat-wake recording instruments have been installedat this location since 1995. Measurements indi-cate that boatwake-induced erosion has beenless here than at river mile 16 (Dorava and

Moore, 1997). The banks near river mile 21 arecovered with grass and some short willowbrush, but the vegetation cover is less densehere than at the downstream site. This site pro-vided an opportunity to test the effects of boat-wakes on a natural unprotected bank and on abank stabilized with an engineered combina-tion of materials.

Site Selection and Description 9

A previously installed, engineered bank-stabilization technique near river mile 21 wastested on September 25, 1998. During the Sep-tember boatwake testing, stream discharge atthe site was about 8,750 cubic feet per second,also based on records at the USGS stream-gag-ing station Kenai River at Soldotna (fig. 1)(Bertrand and others, 1999). This discharge isabout 40 percent less than the long-term meandischarge for August, typically the month ofmaximum mean discharge. This 40 percent dis-charge difference translates into a water depthdifference of about 1.24 feet, indicating that theriver flow during the September test period waswell below mean annual high-water conditions.A segment of bank downstream from the wake-recording instruments had been stabilized witha combination of cabled spruce trees, bio-logs,rocks, and willow plantings (fig. 6) about 2years before the current study. This type ofengineered stabilization is encouraged by rivermanagers as a preferred technique because it isthought to both protect the bank and providefish habitat (Gary Liepitz, Alaska Departmentof Fish and Game, oral commun., 1999). Justupstream from the wake-recording instru-ments, the bank is covered with grass and shortwillows and is in its natural unprotected state.

WAKE HEIGHTS

At the site near river mile 16, a sailingline about 75 feet from the streambank wasestablished with floating buoys. A 20-foot-long, flat-bottom boat, powered by a 28-horse-power outboard, made 20 passes (10 upstreamand 10 downstream) during each of the threetest conditions. The boat was operated at topspeed traveling along the established sailingline with the equivalent weight of three passen-gers on board. A wake gage installed near thebank at this site recorded the maximum wakeheight striking the bank from each boat pass asan ink trace on a paper chart. Initially, data werecollected from the unprotected natural bank,

including the maximum wake heights strikingand the weight of sediment eroded from it dur-ing 20 boat passes.

Following the test on the natural bank, thebio-log was installed in front of the bank (fig.3). This bio-log was not a typical bio-log instal-lation, but rather was intended to simulate theeffect the bio-log would have on boatwakes.Boatwakes approaching the bank struck thebio-log prior to striking the bank or beingrecorded by the wake gage. After this test, thebio-log was replaced by the cabled spruce trees(fig. 4). Boatwakes struck the spruce trees priorto the wake-gage or the streambank. The aver-age of the 20 recorded maximum wake heightswas calculated and used to compare the naturalbank with the same bank after installing thebio-log and the cabled spruce trees (fig. 7, table1).

BANK EROSION

During the wake tests near river mile 16,the average weight of eroded bank material wasmeasured at the natural bank and with eachbank-stabilization technique in place. Thematerial eroded from the streambank duringeach boat pass was collected in a a 12- by 18-by 2-inch baking pan pinned to the river bottomwith the 12-inch side of the edge adjacent to thestreambank. The average weight of erodedbank material retained in the pan during the testwas measured. This measurement was used as

Table 1. Wake heights measured nearriver mile 16, Kenai River, Alaska

Bank-stabilization technique

Wake height (foot)

Average maximum

RangeStandard deviation

None; natural 0.58 0.35—0.83 0.14

Bio-log 0.48 0.25—0.73 0.15

Spruce trees 0.46 0.27—0.76 0.16


Bank Erosion 11

12"

AverageOverhang

AverageUndercut

9"

1.75 FT/FT

6"6"

18"

Bio-logs

Willow & Spruce Tree

2.5 FT/FT

Figure 6. View of site near river mile 21, Kenai River, Alaska: (A) natural streambank and (B) engineered bank stabilization technique.

A

B

THREE GROUPS OF 20 BOAT PASSES AT ONE SITE

0.85

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

0.80

MA

XIM

UM

ME

AS

UR

ED

WA

KE

HE

IGH

T, I

N F

EE

T

Unprotected natural bank Bio-log protected bank Spruce-tree protected bank

Average wake height = 0.48

Average wake = 0.58 height

Average wake = 0.46 height

Figure 7. Height of wakes striking the bank near river mile 16 of the Kenai River, Alaska.

the primary measure of the effectiveness ofeach bank-stabilization technique to retardboatwake-induced streambank erosion. Atriver mile 16, the material eroded from the bankwas collected during each of the 20 boat passes(fig. 8; table 2). A test of bank erosion was alsodone at the site near river mile 21. At this site,the weight of material eroded from the unpro-tected natural bank was compared with mate-rial eroded from the front of the adjacentengineered bank (fig. 6). This material was col-lected during each of nine boat passes (fig. 9;table 2).

Table 2. Weight of eroded bank material retained in collection pan near river miles 16 and 21, Kenai River, Alaska


Average weight (pound per foot of bank)

River mile 16

None; natural 0.00048

Bio-log 0.00014

Spruce trees 0.00010

River mile 21

None; natural 0.163

Engineered restoration 0.044


THREE GROUPS OF 20 BOAT PASSES AT ONE SITE

0

0.0015

0

0.0001

0.0002

0.0003

0.0004

0.0006

0.0007

0.0008

0.0009

0.0010

0.0011

0.0012

0.0013

0.0014

WE

IGH

T O

F E

RO

DE

D B

AN

K M

AT

ER

IAL,

Unprotected natural bank

Average weight = 0.00048

Spruce-tree protected bank


Average weight = 0.00010 IN P

OU

ND

PE

R F

OO

T O

F B

AN

K

Bio-log protected bank

Figure 8. Weight of material eroded from the bank near river mile 16 of the Kenai River, Alaska.

To help understand the accuracy, variabil-ity, and reliability of the weight-measurementmethod, several quality-control and quality-assurance measures were employed. For exam-ple, because the bank-erosion comparisonswere intended to evaluate only boatwake-induced erosion, a sample of bank erosion wasalso taken when no wakes were striking thebank. This type of sample was collected twiceduring each test: before running the boat passesand after all 20 boat passes. The average weightof these bank-erosion samples taken before andafter the boatwake test was then subtractedfrom the average weight of material erodedfrom the banks during each test to determine

the wake-induced erosion. The average weightof material collected at the base of the bankwhen no wakes were striking the bank rangedfrom 5.2 to 41 percent of the average weight ofmaterial eroded during the tests (table 3).

In addition, laboratory proceduresincluded duplicate measurements for 25 per-cent of the samples, to assess for instrumentand human errors in the sample weights. Theerror in laboratory duplicate weights averagedless than 6 percent of the average test sampleweights.

Bank Erosion 13


TWO GROUPS OF NINE BOAT PASSES AT TWO SITES

0

0.38

0

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

0.32

0.34

0.36

IN P

OU

ND

PE

R F

OO

T O

F B

AN

K

Unprotected natural bank


Engineered restored bank


WE

IGH

T O

F E

RO

DE

D B

AN

K M

AT

ER

IAL,

Figure 9. Weight of material eroded from the bank near river mile 21 of the Kenai River, Alaska (see fig. 6A and 6B).

Table 3. Weight of eroded bank material retained in collection pan during, before, and after boatwake tests near river miles 16 and 21, Kenai River, Alaska


Average weight(pound per foot of bank) Percentage of pre- and post-

test average weight eroded during test

During boatwake test Pre- and post-test

River mile 16

None; natural 0.00048 0.000087 18

Bio-log 0.00014 0.000050 37

Spruce trees 0.00010 0.000041 41

River mile 21

None; natural 0.163 -- --

Engineered restoration 0.044 0.0023 5.2

en-sisn-a-dicl,

an is

easeomterity isit--isk2;d

e,

ceorn.g

ter-peatityab-n

aearnkde- a

HABITAT VALUE

The habitat for juvenile chinook salmonthat is provided by each mitigation techniquewas compared, using the amount (area) and thequality (suitability) of habitat at each installa-tion. The amount of habitat provided was deter-mined using standard evaluation procedures(Stiehl, 1994), and habitat quality was evalu-ated using published information on habitatsuitability for juvenile chinook salmon (Ben-dock and Bingham, 1988; Burger and others,1982; Estes and Kuntz, 1986; Liepitz, 1994;and Raleigh and others, 1986).

The most important habitat features forrearing juvenile chinook salmon in the KenaiRiver are generally found along the margin ofthe river. These features are found primarily inareas within about 6 feet of the streambank,where water velocities are less than 1 foot persecond, stream substrate is greater than 1.5inches in diameter, and abundant cover can pro-tect the juvenile chinook salmon from preda-tors. Because the streambank modificationstested did not substantially alter stream sub-strates—at least during the short term—sub-strate size was not considered as a habitatattribute for comparisons. As a result, watervelocity and available protective cover wereselected as the two primary attributes used toevaluate habitat.

Water Velocity

Establishing representative measure-ments of water velocity at the sites was diffi-cult. A single measurement of water velocity ata site may not accurately represent the prevail-ing conditions. In addition, water velocity nat-urally varies considerably in the water columnfrom the bed of the stream where it is a mini-mum to the water surface where it is a maxi-mum. Additionally, the velocity of water alonga river margin also varies substantially. Veloc-ity may be greater near natural or man-made

projections into the channel. It may also bgreater where bank friction is low, and at a miimum where flow friction is greater, such aadjacent to thick vegetation. Water velocity also further complicated by variations in chanel hydraulics. Near streamside bank-stabiliztion projects, water velocity can be alteresubstantially (Dorava, 1995). When a specifstabilization project extends into the channechanges in water velocity can be greater thwhen only the material composing the bankaltered (Dorava, 1995).

During this test, water velocity near thnatural bank at the site near river mile 16 wmeasured with a Price-AA meter. A singlpoint measurement was made about 3 feet frthe bank, at a depth of 0.6 times the total wadepth. In this one location, mean water velocwas about 0.7 foot per second. This velocitywithin the range that results in the highest suability for juvenile salmon. In fact, water velocity ranging from 0.4 to 0.8 foot per second optimal for juvenile chinook salmon (Bendocand Bingham, 1988; Burger and others, 198Estes and Kuntz, 1986; Liepitz, 1994; anRaleigh and others, 1986). Furthermoralthough difficult to verify, it is most likely thatwater velocity near the bio-log and the sprutrees remained within this preferred range fsome distance surrounding their installatioDetermining the area or volume surroundineach bank-stabilization technique where wavelocity is optimal would require three-dimensional velocity measurements beyond the scoof this project. As a result, it was concluded thsingle measurements of mean water velocwould not provide an adequate measure of hitat value from which to compare stabilizatiotechniques.

Another approach to using velocity as measure of habitat value was attempted nriver mile 21 at the site of the engineered bastabilization technique. Near this stabilizebank, multiple mean water velocity measurments were made with a Price-AA meter, at

Habitat Value 15

depth of 0.6 times the total water depth. Thesemeasurements were made at four points at 2-foot increments from the streambank out intothe channel. These multiple velocity measure-ments were then repeated at the adjacent natu-ral bank approximately 20 feet upstream. Theresults indicated that the four mean velocitymeasurements made at 2, 4, 6, and 8 feet fromthe bank averaged 0.889 foot per second at thestabilized bank and 0.869 foot per second at thenatural bank. This small difference in velocityis essentially insignificant to juvenile chinooksalmon, providing suitability indices of 91 and93 percent of optimum values, respectively(Bendock and Bingham, 1988; Burger and oth-ers, 1982; Estes and Kuntz, 1986; Liepitz,1994; and Raleigh and others, 1986).

This exercise indicates that even moredetailed measurements of water velocity can bedifficult to use for habitat value comparison. Asa result, water velocity was not used in thisstudy as an attribute to compare bank-stabiliza-tion techniques. Rather, water velocity wasused only as a qualitative comparison; valueswere assumed to fall within the range of opti-mum suitability somewhere near each site, atleast during the short period of testing.

Protective Cover

Protective cover has been identified as aessential component of chinook salmon rearinghabitat by several researchers (Bendock andBingham, 1988; Burger and others, 1982; Estesand Kuntz, 1986; Liepitz, 1994; Raleigh andothers, 1986). The protective cover for juvenilechinook salmon provided by natural andrestored banks was evaluated at both studysites. Near river mile 16, the amount of protec-tive cover provided by the bio-log and by thespruce trees was compared directly. Both thebio-log and spruce trees provide protectivecover to juvenile chinook salmon that can hideunder them. The amount of habitat provided bythe bio-log was limited to the diameter of the

log, about 12 inches. The amount of habitatprovided by the cabled spruce trees includedboth its 20-foot length and the extension of thelimbs into and above the water. The bio-logprovides an area coverage of about 1 foot perlinear foot (table 4). In contrast, the two cabledspruce trees form a triangular-shaped protec-tive area about 6 feet wide at the base. Becausejuvenile chinook salmon can find protectivecover anywhere under the trees, the entire areaunder the trees provides cover. The area of thistriangular-shaped space can be determined asone-half the width of the base of the trees, mul-tiplied by the 20-foot length. This computationresults in an average of about 3 feet of availableprotective cover per linear foot of tree providedby the spruce trees (table 4). This calculationindicates that the tested cabled spruce-tree sta-bilization technique potentially provided aboutthree times more cover to juvenile chinooksalmon than the tested bio-log.

Near river mile 21, the habitat provided asprotective cover by the stabilized bank wascompared with that provided by the adjacentnatural bank. During bankfull conditions, thenatural bank had about a 9-inch-deep undercutat the base of the bank and was covered prima-rily by grass vegetation that overhung about 12

Table 4. Amount of protective cover for juvenilesalmon at river miles 16 and 21, Kenai River, Alaska


Protective cover(foot per linear foot)

River mile 16

None; natural 0 base

Bio-log 1.0

Spruce trees 3.0

River mile 21

None; natural 1.75

Engineered restoration 2.50


r-nk,ndendne

ve

c-atwt-

e

lesser-osi-efnknd5

raloftly

fna-e-ankro-isatatr

d a

hattere-

inches over the river (fig. 6A). This bank con-figuration resulted in an average of about 1.75feet of cover per linear foot of bank (table 4).About 20 feet downstream at the stabilizedbank, the willow plantings and cabled sprucetrees both extended into the water about 18inches (fig. 6B). Additionally, the bio-logs atthe bank base extended streamward about 6inches in a two-layered tier. This arrangementcreated three layers of protective cover, result-ing in about 2.5 feet of protective cover per lin-ear foot of bank (table 4).The comparisons atthis site indicate that restoration efforts createdmore cover than the adjacent natural bank.However, it should be noted that each restora-tion project is different and the cost of creatingsuitable juvenile chinook salmon habitat iscommonly greater than property owners wantto expend. In addition, because each naturalbank is different and each may provide more orless potential habitat, making site-to-site com-parisons is difficult.

DISCUSSION

Using a two-sample t-test (α = 0.05; Zar,1984), the average maximum wake heightsstriking the unprotected natural bank were sig-nificantly greater than those striking the bankwhere it was protected by a bio-log or cabledspruce trees. Where the bank was protected byspruce trees, average values for maximumwake heights were not significantly different (α= 0.05) from where the bank was protected bythe bio-log. The average maximum wakeheights were not significantly differentbetween bio-log and spruce tree bank-stabiliza-tion techniques. Both techniques share charac-teristics that help reduce wake heights strikingthe banks, such as materials extended above thewaterline and into the stream. These types ofmaterials will intercept boatwakes and, conse-quently, attenuate boatwakes. The wake-atten-uation characteristics of bio-logs and sprucetrees shown by this study’s data set are also

supported by engineering intuition. Interfeence with boatwakes as they approach the baas provided by spruce-tree limbs that extesubstantially into and above the water, will beffective at a wide range of water depths acan be adopted into other bank-stabilizatiotechniques. For example, a bio-log could battached to the streambed and bank with liwillow stakes that later will grow into maturewillows that enhance wake-attenuating charateristics. In addition, submerged vegetation thhas either fallen into the water or grows belothe waterline would most likely attenuate boawakes as they approached the streambank.

Reducing the height of wakes striking thbank will most likely lead to a reduction inbank erosion, because smaller wakes have energy to dissipate on the streambank (Andson, 1975; Dorava and Moore, 1997; Limerinand Smith, 1975; U.S. Army Corps of Engneers, 1984; von Krusenstierna, 1990; Yousand others, 1978). The measured streambaerosion at banks protected by spruce tree abio-logs was significantly reduced to at least 2percent of what it was at the unprotected natubank. The erosion-control effectiveness spruce trees and bio-logs was not significandifferent at an α level of 0.05 (Zar, 1984).

At river mile 21, where one segment obank was protected by the engineered combition of materials, the boatwake-induced strambank erosion measured at the protected bwas 27 percent of that measured at the unptected natural bank. The characteristics of thengineered bank-stabilization technique threduce bank erosion are similar to those thwere effective at the river mile 16 site. Foexample, the engineered stabilization site habio-log toe with willow shrubs growingthrough it, and a layer of spruce trees above t(fig. 6B). Bank-stabilization techniques thainclude projections into and above the watwill reduce both wake height and boatwakinduced streambank erosion.

Discussion 17

Bio-logs are a relatively new erosion con-trol technique, whereas the low-cost erosionreduction provided by spruce trees has beenrecognized for some time. An illustrativeexample is the use of Christmas trees as erosioncontrol along the coastline of Louisiana (Bahl-inger, 1996; Dunne, 1999) (fig. 10).

The comparisons of habitat value or hab-itat suitability provided by each tested stabili-zation technique indicated that cabled sprucetrees potentially provided more cover for juve-nile chinook salmon than bio-logs did. Thecharacteristics of the spruce trees that help toattenuate wakes and reduce bank erosion alsoprovided valuable cover to juvenile chinook

salmon. The habitat provided by the engineeredbank-stabilization project at river mile 21 wasgreater than that provided by the adjacent natu-ral bank. When selecting a type of bank stabili-zation, the relative contribution to fish habitatmay be a valuable consideration in addition toeffectiveness at reducing wake heights andbank erosion. Another consideration for selec-tion is the durability of the stabilization tech-nique. For example, as spruce trees age, thepossible loss of needles and limbs may reducethe ability of this technique to attenuate wakesand provide fish habitat. However, determiningthe potential reduction in effectiveness of agingspruce trees or their anticipated life span isbeyond the scope of this study.


eof abi-atat

todat-hethee-

di-e-nes,dn

ebi-u-

s-agei-2,

elri-

on in7-i-ly

Although these tests of bank-stabilizationtechniques indicate an enhancement of fishhabitat and an effective reduction in erosionfrom the tested techniques, these tests wereconducted during specific conditions. As aresult, no information is provided about howwell the bank-stabilization techniques performover longer periods. For example, it is not clearwhich technique is more effective at reducingflood-induced erosion, which can occur infre-quently—every 10 to 100 years. In addition, noevaluation has been made to compare the effec-tiveness of the bank-stabilization techniquesduring different climatic conditions such as fol-lowing major rainfalls or extensive periods ofno rainfall, when streambanks might react dif-ferently to boatwakes. With these limitations,the results of this study indicate the relativeeffectiveness of several popular bank-stabiliza-tion techniques in use along the Kenai River.The study also provides insight as to how tomodify or adapt other streambank-stabilizationtechniques specifically to protect shorelinesfrom boatwake-induced streambank erosion.

Managers of the Kenai River or other riv-ers subjected to boat traffic can use this infor-mation when identifying preferred streambank-stabilization techniques. Additionally, stream-side property owners can use this informationwhen selecting a method to protect their riverfrontage from boatwake-induced erosion.

SUMMARY

This study of the effectiveness of stream-bank-stabilization projects along the KenaiRiver was motivated by the economic impor-tance of the Kenai River fisheries and thepotential threat to these fisheries from boat-wake-induced streambank erosion. The manag-ers of the river are concerned that currentstreambank restoration practices may not beadequately addressing boatwake-induced stre-ambank erosion, which generates undesirablenon-point source pollution or damages impor-

tant riparian and riverine fish habitat. Thresults of this study provide a comparison three bank-stabilization techniques anddescription of some characteristics of the stalization techniques that are most effective reducing boatwake-induced erosion and thprovide protective cover for juvenile salmon.

Controlled in-river tests were designed evaluate the ability of a few commonly usebank-stabilization techniques to attenuate bowakes and retard streambank erosion. Tresults indicated that spruce trees cabled to streambank and bio-logs pinned to the strambed significantly reduced both maximumwake heights and streambank erosion. In adtion, spruce trees provided more protectivcover for juvenile chinook salmon than biologs did. An engineered bank-stabilizatioproject that used a combination of spruce trebio-logs, rocks, and willow plantings providemore protective cover and erosion protectiothan the adjacent unprotected natural bank.

REFERENCES CITED

American Society of Civil Engineers, 1992, Sedimentand aquatic habitat in river systems—Task forccommittee on sediment transport and aquatic hatats. Sedimentation Committee: Journal of Hydralics, v. 118, no. 5, p. 669-687.

Anderson, F. E., 1975, The short term variation in supended sediment concentration caused by passof a boat wave over a tidal flat environment: Unversity of New Hampshire, Technical Report No. 40 p.

Bahlinger, K.D., 1996, Effectiveness of Christmas treprojects: Louisiana Department of NaturaResources, PCWRP Information Series No. 2, vaously paged.

Bendock, T., and Bingham, A., 1988, Juvenile salmseasonal abundance and habitat preferenceselected reaches of the Kenai River, Alaska, 1981988: Alaska Department of Fish and Game, Divsion of Sport Fish, Fishery Data Series 70, variouspaged.

Summary 19

tatd--

sd

81-s-

ntr:

at-

p.

kg

ut onst

t:

tsel

fn--

edero31t.

.,-

e-rch

veishf

Bentley, Cliff, Parke, Gary, and Leuthe, David, 1991,Streambank erosion—Gaining a greater under-standing: Minnesota Department of NaturalResources, Division of Water, 33 p.

Bertrand, C.M., Hess, D.L., Meyer, D.F., Jackson. M.L.,and Smith, C.W., 1999, Water resources data forAlaska—Water year 1998: U.S. Geological SurveyWater-Data Report 98-1, 321 p.

Beschta, R.L., 1989, The intrusion of fine sediments intoa stable gravel bed: Journal of the FisheriesResearch Board of Canada, v. 36, no. 2, p. 204-210.

Bhowmik, N.G., and Demissie, M., 1982, Waves gener-ated by river traffic: Conference on ApplyingResearch to Hydraulics Practice, American Societyof Civil Engineers Hydraulic Division, Proceed-ings, p. 179-187.

Bjorn, T.C., 1969, Embryo survival and emergence stud-ies: Idaho Fish and Game Department Job Compli-ance Report, Project F-49-R-7, 11 p.

Burger, C.V., Wangaard, D.B., Wilmot, R.L., and Palm-isano, A.N., 1982, Salmon investigations in theKenai River, Alaska, 1979-1981: U.S. Fish andWildlife Service National Fisheries Research Cen-ter, 139 p.

Byrd, J.E., and Perona, M.J., 1980, The temporal varia-tions of lead concentration in a freshwater lake:Water, Air and Soil Pollution, v. 13, p. 207-220.

Das, M.M., and Johnson, J.W., 1970, Waves generatedby large ships and small boats: Twelfth CoastalEngineering Conference, Proceedings, v. 3, p.2281-2286.

Dezman, L.E., 1990, Can erosion control meet regula-tory standards?: National Conference, AmericanSociety of Civil Engineers Hydraulics Division,Chang, H.H. and Hill J.C. ed., Proceedings, p. 391-396.

Dorava, J.M., 1995, Hydraulic characteristics nearstreamside structures along the Kenai River,Alaska: U.S. Geological Survey Water-ResourcesInvestigations Report 95-4226, 41 p.

Dorava, J.M., and Moore, G.W., 1997, Effects of boat-wakes on streambank erosion, Kenai River, Alaska:U.S. Geological Survey Water-Resources Investi-gations Report 97-4105, 84 p.

Duijsings, J.J.H.M., 1986, Streambank contribution tothe sediment budget of a forest stream: Amsterdam,Universiteit, Fysisch-Geografisch en BodemkundigLaboratorium, v. 40, 190 p.

Dunne, Mike, 1999, Christmas trees take on tidal-lakemission: Lafayette Louisiana, The Advocate, Janu-ary 27, 1999, p. 2B.

Estes, C.C., and Kuntz, K.J., 1986, Kenai River habistudy: Alaska Department of Fish and Game Feeral Aid in Fish Restoration F-10-1 and Anadromous Fish Studies, v. 27, variously paged.

Garcia, K.T., 1988, Effect of erosion-control structureon sediment and nutrient transport, EdgewooCreek drainage. Lake Tahoe basin, Nevada, 1983: U.S. Geological Survey Water-Resources Invetigations 87-4072, 65 p.

Garrad, P.N., and Hey, R.D., 1987, Boat traffic, sedimeresuspension and turbidity in a Broadland riveJournal of Hydrology, v. 95, p. 289-297.

Garrad, P.N., and Hey, R.D., 1988, The effect of botraffic on river regime: Proceedings of the International Conference on River Regime, Paper J1,395-409.

Hagerty, D.J., Spoor, M.F., and Ullrich, C.R., 1981, Banfailure and erosion on the Ohio River: EngineerinGeology, v. 17, p. 141-158.

Hansen, E.A., 1971, Sediment in a Michigan trostream—Its source, movement, and some effectsfish habitat: U.S. Department of Agriculture ForeService Research Paper NC-59, 14 p.

Hilton J,. and Phillips, G.L., 1982, The effect of boaactivity on turbidity in a shallow Broadland riverJournal of Applied Ecology, v. 19, p. 143-150.

Jackivicz, T.P., and Kuzminski, L.N., 1973a, The effecof the interaction of outboard motors with thaquatic environment—A review: EnvironmentaResearch, v. 6, p. 436-454.

Jackivicz, T.P., and Kuzminski, L.N., 1973b, A review ooutboard motors effects on the aquatic enviroment: Journal of Water Pollution Control Federation, v. 45, no. 8, p. 1759-1770.

Karaki, S. and vanHoften J., 1974, Resuspension of bmaterial and wave effects on the Illinois and UppMississippi Rivers caused by boat traffic: ColoradState University Engineering Research Center, p. [Prepared for U.S. Army Engineer District, SLouis.]

Klingeman, P.C., Matin, Habibollah, and Huang, C.-C1990, Investigation of motorboat-induced streambank erosion on the Lower Deschutes River: Orgon State University, Water Resources ReseaInstitute Report, 67 p. + appendixes.

Liepitz, G.S., 1994, An assessment of the cumulatiimpacts of development and human uses on fhabitat in the Kenai River: Alaska Department oFish and Game Technical Report No. 94-6, 63 p.


,e-in:

veoat

k--

elan,5-

esi-

ro-n-ly

onchp.ly

al9,

edi-geo-es

yer,-

,e-.

onet

er-

Lietman, P.L., Ward, J.R., and Behrendt, T.E., 1984,Effects of specific land uses on nonpoint sources ofsuspended sediment, nutrients, and herbicides—Pequea Creek basin, Pennsylvania, 1979-80: U.S.Geological Survey Water-Resources Investigations83-4113, 88 p.

Limerinos, J.T., and Smith, Winchell, 1975, Evaluationof causes of levee erosion in the Sacramento-SanJoaquin Delta, California: U.S. Geological SurveyWater-Resources Investigations 28-74, 53 p.

Meehan, W.R., 1974, The forest ecosystem of southeastAlaska. Part 3, Fish habitats: U.S. Forest ServiceGeneral Technical Report PNW-15, 15 p.

Meehan, W.R., and Swanston, D.N., 1977, Effects ofgravel morphology on fine sediment accumulationand survival of incubating salmon eggs: U.S. ForestService Research Paper PNW-220, 16 p.

Murphy, K.J., and Eaton, J.W., 1983, Effects of pleasure-boat traffic on macrophyte growth in canals: Journalof Applied Ecology, v. 20, p. 713-729.

Nanson, G.C., Von Krusenstierna, A., Bryant, E.A., andRenilson, M.R., 1994, Experimental measurementsof river-bank erosion caused by boat-generatedwaves on the Gordon River, Tasmania: RegulatedRivers—Research and Management, v. 9, p.1-14.

Nelson, L.M., 1986, Effect of bank protection measures,Stehekin River, Chelan County, Washington: U.S.Geological Survey Water-Resources InvestigationsReport 85-4316, 22 p.

Odgaard, A.J., 1984, Bank erosion contribution to streamsediment load: Iowa Institute of Hydraulic ResearchReport No. 280, 92 p.

Raleigh, R.F., Miller, W.J., and Nelson, P.C., 1986, Hab-itat suitability index models and instream flow suit-ability curves—Chinook salmon: U.S. Fish andWildlife Service, Biological Report 82 (10.122), 59p.

Reckendorf, F.F., 1991, Streambank erosion: problems,causes, and solutions, Fifth Federal InteragencySedimentation Conference Proceedings, p. 363-370

Schoellhamer, D.H., 1990, Observation of sedimentresuspension in Old Tampa Bay Florida: NationalConference, American Society of Civil EngineersHydraulics Division, Chang, H.H. and Hill J.C.eds., Proceedings, p. 51-56.

Scott, K.M., 1982, Erosion and sedimentation in theKenai River, Alaska: U.S. Geological Survey Pro-fessional Paper 1235, 35 p.

Smart, M.M., Rada, R.G., Nielsen, D.N., and ClaffinT.O., 1985, The effect of commercial and recrational traffic on the resuspension of sediment Navigation Pool 9 of the Upper Mississippi RiverHydrobiologia, v. 126, p. 263-274.

Sparks, R.E., 1975, Possible biological impacts of wawash and resuspension of sediments caused by btraffic in the Illinois River: Illinois Natural HistorySurvey, 23 p.

Stiehl, R.B., 1994, Habitat evaluation procedures worbook: Fort Collins Colo., National Biological Survey, variously paged.

Thomas, W.A., 1990, Example of the stable channdesign approach: National Conference, AmericSociety of Civil Engineers Hydraulics DivisionChang, H.H. and Hill J.C. eds., Proceedings, p. 17180.

Trimble, S.W., 1997, Streambank fish-shelter structurhelp stabilize tributary streams in Wisconsin: Envronmental Geology, v. 32, no. 3, p. 230-234.

U.S. Army Corps of Engineers, 1981, Streambank esion demonstration project: Final Report to Cogress, December 1981, Volume 1 of 2, variouspaged.

U.S. Army Corps of Engineers, 1984, Shore protectimanual (4th ed.): Coastal Engineering ResearCenter, Waterways Experiment Station, v. 1, cha1-5, and v. 2, chap. 6-8 + appendixes, variouspaged.

Verdon, Michael, 1999, Can the ICW (IntraCoastWaterway) be saved?: Boating World, March 199p. 47-51.

von Guerard, P., 1989, Suspended sediment and sment-source areas in the Fountain Creek DrainaBasin upstream from Widefield, southeastern Colrado: U.S. Geological Survey Water-ResourcInvestigations Report 88-4136, 36 p.

Von Krusenstierna, Axel, 1990, River bank erosion bboat-generated waves on the lower Gordon RivTasmania: The University of Wollongong, Australia, M.S. thesis, 136 p.

Yousef, Y.A., McLellon, W.M., Fagan, R.H., ZebuthH.H., and Larrabee, C.R., 1978, Mixing effects duto boating activities in shallow lakes: Florida Technological University, Technical Report ESEI No78-10, 352 p.

Zabawa Chris, and Ostrom Chris, 1980, Final report the role of boat wakes in shore erosion in AnnArundel County, Maryland: Maryland Departmenof Natural Resources, variously paged

Zar, J.H., 1984, Biostatistical analysis (2d ed.): New Jsey, Prentice Hall, 718 p.

References Cited 21

effectiveness of streambank-stabilization techniques along ...effectiveness of...

Documents