UNITED STATES DEPARTMENT OF THE INTERIORStewart L. Udall, Secretary

James K. Carr, Under SecretaryFrank P. Briggs, Assistant Secretary for Fish and Wildlife

FISH AND WILDLIFE SERVICE, Clarence F. Pautzke, Commissioner

BUREAU OF COMMERCIAL FISHERIES, Donald L. McKernan, Director

I

SUCCESS OF PINK SALMON SPAWNING RELATIVE

TO SIZE OF SPAWNING BED MATERIALS

by

William J. McNeil and Warren H. Ahnell

Contribution No. 157, College of Fisheries, University of Washington

United States Fish and Wildlife ServiceSpecial Scientific Report-Fisheries No. 469

Washington, D.C.January 1964

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Field measurement of size composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Collecting samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sorting and measuring samples ......................................

Permeability and its relation to composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Method of measuring permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Results of permeability tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Size composition and spawning success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

......................................Silt removal by spawners. . . . . .

Volumes removed . . . . . . . . .

Organic content. . . . . . . . .

Spawning bed siltation. . . . . . . .

Silt pollution from logging. . l .

Silt content of spawning beds

Effect of silt on permeability

Silt removal by flooding . . . . .

Summary . . . . . . . . . . . . . . . .

Literature cited . . . . . . . . . . . .

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

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

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

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

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

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

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

. . . . . ..-..............................

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

6

9

9

9

9

9

12

12 <

12 :

14

1 4

iii

i

SUCCESS OF PINK SALMON SPAWNING RELATIVETO SIZE OF SPAWNING BED MATERIALS

by

William J. McNeil, Research Associateand

Warren H. Ahnell, Research AssistantFisheries Research Institute, College of Fisheries

University of Washington, Seattle, Washington

ABSTRACT

The potential of a salmon spawning bed to produce fry is directly related to itspermeability. The relationship between the coefficient of permeability and the frac-tion of bottom materials consisting of fine particles is inverse.

Field methods for measuring size composition of bottom materials in salmonspawning beds are described, and an empirical relationship between the fraction(by volume) of solids less than 0.833 mm. minimum dimension and the coefficientof permeability of stream bottom materials is given. Size of bottom materials instreams utilized for spawning by pink salmon (Oncorhynchus gorbuscha) varied con-siderably. The more productive spawning streams had the more permeable spawningbeds. Adult pink salmon caused the removal of finer particles from bottom mate-rials during spawning. The evidence indicates that the fine particles removedconsist largely of organic matter. Logging caused fine sands and silts to accrueto spawning beds. Flooding caused the removal of fine particles from spawning beds.

,

.

INTRODUCTION

Pink salmon (Oncorhynchus gorbuscha) arethe most abundant of the Pacific salmon andin most years provide a larger commercialcatch than the other species. In the easternPacific, pink salmon are of commercial im-portance from Bristol Bay, Alaska, to PugetSound, Wash. They are of greatest importancein Southeastern Alaska where there are about1,100 spawning streams (Martin, 1959).

Note--William J. McNeil is now with the Bureau ofCommercial Fisheries, Auke Bay, Alaska; Warren H.Ahnell is now at Washington State University, Pullman,Wash.

A critical period in the life history of pinksalmon occurs between the time eggs are de-posited in spawning beds of streams and thetime fry emerge several months later. Adultfemales excavate pockets in gravel beds andcover their eggs with 3 to 15 inches of gravel.This action affords eggs and larvae protectionagainst predators, mechanical injury, anddisplacement by flowing water, Other environ-mental stresses are encountered, however,and mortality within the streambed commonlyexceeds 75 percent.

The growth, development, and survival ofsalmon eggs and larvae are dependent onphysical and chemical characteristics of thesurrounding water. These properties include

temperature, dissolved oxygen content, veloc-ity, mineral and waste metabolite content,and osmotic pressure. Osmotic pressure isespecially important where spawning occur6in intertidal area6 of streams.

To prosper, an embryo or larva must re-ceive an ample supply of oxygenated waterof suitable temperature and free of toxicsubstances.’ Thus, the quality of water withina spawning bed may limit the number of salmonproduced. Size composition of bottom material6greatly influences water quality by affectingrates of flow within spawning bed6 and rate6 ofexchange between intragravel 1 and streamwater.

The presence of fine particle6 in spawningbeds (viz, sands and silts) increases egg andlarval mortality of several salmonid species(Harrison, 1923; Shapovalov, 1937; Shaw andMaga, 1943; Shelton, 1955; Lucas, 1960; Andrewand Geen, 1960; Cordone and Kelley, 1961).The presence of fine materials in spawningbeds reduces their permeability, and accordingto Wickett (1958), survival of pink and chum(0. ke ta) salmon eggs and larvae is related di-rectly to permeability.

Equipment ha6 been developed to measurepermeability of salmon spawning beds in situ(Terhune, 1958). There is a possibility thatthe size composition of bottom materials inspawning beds affords an equally satisfactoryindex of streambed quality a6 it relates toegg and larval survival. It is the purpose ofthis report to (1) describe a method for meas-uring size composition of bottom materials insalmon spawning beds, (2) show a relationshipbetween content of fine materials and per-meability of spawning beds, (3) describe dif-ferences in size composition and permeabilityof spawning beds in streams supporting low tohigh densities of spawning adult pink salmon,and (4) demonstrate the occurrence of tem-poral variations in size composition associatedwith spawning, logging, and flooding.

The size composition of bottom materialswas studied as part of an investigation of the

1 The term “intragravel water” is used to describewater occupying interstitial spaces within the streambed.

effects of logging on pink salmon streams: inAlaska. Observations were made mostly instreams located near Hollis on Prince of WalesIsland. Financial support for these studies wasprovided by the Bureau of Commercial Fish-eries, U.S. Fish and Wildlife Service, withSaltonstall-Kennedy Act funds. Assistance withfield sampling studies was given by the North-ern Forest Experiment Station, U.S. ForestService, Juneau.

FIELD MEASUREMENT OF SIZECOMPOSITION

Perhaps the simplest technique for deter-mining size composition of spawning bedmaterials was employed by Burner (1951),who approximated visually the relative amountsof large, medium, and small gravel in redds,A more analytical classification of gravels inPacific salmon redds was undertaken by Cham-bers, Allen, and Pressey (1955), who isolatedsmall areas of stream bottom from the in-fluence of current with open cylinders. Gravelwas removed from within the cylinders witha scoop, washed through a series of sieves,and weighed. Hatch (1957) collected bottommaterials with a spade that had a hood enclos-ing the upper section of blade. In a laboratory,he analyzed bottom materials for sand, silt,and clay content.

Collecting Samples

We developed a bottom sampler suitablefor use in shallow water and in gravel afterseveral types of samplers were tried andfound to have a common deficiency--noneretained silt going into suspension. Further-more, samplers with mechanically operatedclosures did not function successfully at alltimes in coarse gravel and rubble, and certainsampling techniques required the collection oflarge quantities of bottom materials that couldnot be sorted and classified quickly.

Samples of bottom materials were removedfrom the streambed with the sampler (fig. 1).The sampler is stainless steel and is roundin cross section. The tube of the sampler wasworked manually to a depth of 6 inches. Con-tents of the tube were dug by hand and lifted

Figure

into

into awas cthe ticap and

Th.bias

thestreaabsolt h e lecteTo nwas

Th

A 4

gravdiamfrom10-qAfte:suspwas

2

ini in1lesINilS

.sh-withwithrth-rest

E

:er-bed511,antsIds.s inam-2tedin-

rive1vithres,tomios-)=Y,silt,

ablefterand

lonerer-ated

alltainn of)uld

)ved01).)undwas:on-fted

,’.

Water surface

( Silt in suspension)

Handle Sorting and Measuring Samples

Cap

Basin

.Gasket

Figure 1..-Sampler for collecting bottom materials.

into the basin. Because the water was con-tinuously agitated, the finest materials passedinto and remained in suspension as the samplewas collected. After solids were removed fromthe tube and placed in the basin, a watertightcap was inserted in the tube to retain waterand suspended solids in the basin.

This technique possibly introduced somebias because of the loss of silt in suspensionwithin the tube as the sampler was lifted fromthe stream. Samples collected during lowstream discharge no doubt contained a greaterabsolute quantity of solids in suspension withinthe tube, on the average, than samples col-lected during periods of high stream discharge.To minimize this source of error, samplingwas restricted to periods of low discharge.

The diameter of the tube used at a particularlocation depended on the size of the gravel.A &inch-diameter tube was used in moststreams sampled. In a stream with coarsegravel it was necessary to use a 6-inchdiameter tube. After a sample was removedfrom the streambed, it was transferred tolo-quart plastic buckets to facilitate handling.After at least 10 minutes had elapsed to allowsuspended materials to settle, excess waterwas decanted from the buckets.

In Southeastern Alaska, where it rainsfrequently, the problem of drying samplesbefore sorting is almost insuperable unlesscovered facilities are provided. For sortingsamples in the field, it was necessary there-fore to adopt "wet" techniques, and the sampleswere not dried for analysis.

Bottom samples were separated into 10 sizeclasses by washing and shaking them throughnine standard Tyler sieves having the follow-ing square mesh openings (in mm.): 26.26,13.33, 6.68, 3.33, 1.65, 0.833, 0.417, 0.208,and 0.104. Silt passing the finest screen wascollected in a vessel.

Washing samples through sieves is efficientand affords nearly complete separation of sizegroups. One problem with very fine sieves,however, is their tendency to become cloggedand to retain water. The finest sieve used(0.104 mm.) was found to be the smallest thatwould allow water and suspended materials topass through without seriously clogging thesieve.

After passing through the finest screen,water and suspended materials were placedin a large funnel (fig. 2) and allowed to settlefor 10 minutes. The volume of settled solidswas then measured. Solids remaining in sus-pension were discarded. More than 90 percentof the suspended solids settled within 10minutes.

The volume of solids retained by each sievewas measured after the excess water drainedoff. The contents of each of the larger sieves(26.26 through 0.417 mm.) were placed in thedevice shown in figure 3, and the water dis-placed by solids was collected in a graduatedcylinder and measured. Solids retained by thesmall diameter sieves (0.208 and 0.104 mm.)were sometimes carefully washed into a gradu-ate with a measured volume of water, and theincrease in volume was read directly. In theseinstances, the fines were transferred to thegraduate through a 6-inch plastic funnel.

3

Water surface

\b- Rubber tubing

125ml. test tube

Sil t

Figure 2 . Settling funnel for collecting silt fraction inbottom materials.

Two men could sort a sample and measurethe volume of solids retained on each sievein about 10 minutes.

The volume of individual samples collectedwith the bottom sampler varied somewhatfrom point to point but was generally within10 percent of the mean. Variation in samplevolumes was caused by variation in porosityand core depth. All sample fractions areexpressed as a percentage of the sample.

Graduatedisplacedsieve con

by the ztent

z=

surface

-

Figure 3.--Device for measuring the volume of waterdisplaced by solids retained by sieves.

PERMEABILITY AND ITSRELATION TO COMPOSITION

The permeability of spawning bed materialsdepends on the density and viscosity of thewater, the porosity of the streambed, and thesize, shape, and arrangement of solids. Thereis a direct relationship between permeabilityand porosity (Franzini, 1951) and betweenpermeability and average particle size inmixtures of nonuniform sizes (Krochin, 1955;Childs, Collis-George, and Holmes, 1957).A decrease of temperature and porosity and analteration of particle shape from spherical toangular produce marked reductions in per-meability of sands (Fair and Hatch, 1933).

A standpipe for measuring permeability ofsalmon spawning beds has been tested andcalibrated (Pollard, 1955; Terhune, 1958).Wickett (1958) used this permeability stand-pipe to measure permeability of spawningbeds in British Columbia streams wheretotal fresh-water survival had been measuredover a number of years. Wickett observed arelationship between average permeability ofthe beds and survival of salmon (fig. 4),

4

Fj(

1

fZP’tiPIailf

a ,

flnfl

S

nntih

C

--

tface

legs

3ter

:alsthetheerelityeen.

5;9.lani toZr-;31.

o fmda.Id-ingtre‘ed;Zaof

0 40 80 120 160Permeability of streambed gravels (cm./min.)

Figure 4--Observed relationships reported by Wickett(1958) between permeability of spawning beds and sur-vival of pink and chum salmon to the migrant fry stage.

Method of Measuring Permeability

The coefficient of permeability describes thefacility with which a liquid can flow throughpermeable materials. Since viscosity affectsthe rate at which a liquid flows, it is commonpractice to express permeability at a temper-ature selected as a standard (68O F. was se-lected in these studies).

The coefficient of permeability (k) is definedas

k=+-.

where Q is the volume of water flowing perunit of time through a gravel bed having ahydraulic gradient i and a cross-sectional areaA. Since i is dimensionless, the dimensionsof k are

lengthtime l

Samples of bottom materials were collectedfrom two streams. Each sample consisted oftwo 6-inch-diameter cores that were removedfrom adjacent points in important pink salmonspawning areas. Size composition was deter-mined, and each sample was thoroughly re-mixed and divided into 12 fractions. The frac-tions were added one at a time to a constant-head permeameter (fig. 5).

To attain a relatively uniform degree ofcompaction (and porosity), gravel was neither

Q measuredat outlet

tbh

I_

I -i

Figure 5-Constant-head permeameter used to meas-ure permeability ofspawningbedmaterials. Hydraulicgradient (i)= A h/l.

tamped nor shaken after being placed in apermeameter. H e n c e, the permeabilitiesmeasured were most likely representative ofloose, unconsolidated spawning bed materials.

The coefficient of permeability was meas-ured for 19 bottom samples. Water was allowedto pass through each sample continuously overthe period August 25-30, 1960. Permeabilityof each sample was measured August 26, 29,and 30. Water temperature varied from 490 to56’ F. at the times measurements weremade,and all permeability readings were correctedfor standard viscosity of 1.0 centipoise at680 F.*

Results of Permeability TestsA good inverse relationship was found be-

tween the coefficient of permeability and the a.c

* Each observed value of permeability was multiplied ’ :,, flby water viscosity (in centipoises) at the temperature , /. *a measurement was made to obtain permeability at 680 . d, )F. (see Frevert and Kirkham, 1948). . *, *, ., .

5. ., .. *# .. ., *. I. *

,

1 ’

,

r

.

1

I

Table 1 . - -Percentage of bottom materials passing through 0.833.mm. sieve andcoefficient of permeability of 19 bottom samples measured on three dates

[Permeability readings corrected for standard viscosity of 1.0 centipoise at 680 F.]

P e r c e n t a g epassing through August 26 August 29 August 30 Mean0.833-mm. s i eve

Cm. /min. Cm. /min. Cm. /min. Cm./min.4 . 5 461 5 5 6 512 5105.0 363 381 4 6 0 4015 . 5 5 4 4 4 7 6 5 3 8 5 1 95 . 6 291 2 9 0 2 8 3 2 8 85 . 6 370 281 2 8 8 3136 . 1 2 5 2 2 7 8 2 8 0 2 7 06 . 2 2 8 8 3 6 3 3 4 4 3 3 26 . 3 3 0 5 2 4 3 315 2 8 88 . 1 159 229 158 1829 . 3 150 1 9 4 196 1809 . 7 144 188 156 163

10.1 152 2 5 6 172 1951 0 . 3 168 2 2 3 141 1771 0 . 9 5 3 60 61 5 81 2 . 2 131 128 134 1311 2 . 7 4 2 4 5 4 3 4 31 2 . 7 8 7 105 1 0 4 9 91 4 . 7 2 5 21 4 1 2 915.6 2 4 2 6 2 3 2 4

percentage by volume of a bottom samplepassing through an 0.833.mm. sieve. Per-centage of materials passing through an 0.833.mm. sieve are given in table 1 along with theobserved values of the permeability coefficient.Mean permeability is plotted against per-centage of materials passing through an 0.833.mm. sieve in figure 6.

It is apparent that permeability is highwhere bottom materials contain less than 5percent by volume of sands and silts passingthrough an 0.83%mm. sieve. Low permeabilityoccurs where bottom materials contain morethan 15 percent by volume of sands and siltspassing through an 0.833.mm. sieve.

SIZE COMPOSITION ANDSPAWNING SUCCESS

Bottom samples were obtained frompling areas in six streams. Most of the

sam-effort

6

was expended in five streams near Hollis(Old Tom, Maybeso, Indian, and TwelvemileCreeks, and Harris River). Samples also werecollected from Anan Creek, which is locatedabout 75 miles northeast of Hollis and is oneof the most productive pink salmon streamsin Southeastern Alaska. 3

All six streams were sampled in summer1959 shortly before spawning began. Table 2gives the mean percent by volume of solidsretained by sieves and settling from suspen-sion in 10 minutes. Pronounced differenceswere found in volumes of fine sands and siltsin the various spawning beds (fig. 7).

3 For estimates of spawning escapements see "StreamCatalog of Southeastern Alaska Regulatory District No.2” (Special Scientific Report--Fisheries No. 453) and“Stream Catalog of Southeastern Alaska, RegulatoryDistricts Nos. 5, 6, 7, and 8” (in preparation at theFisheries Research Institute for publication as SpecialScientific Report-Fisheries).

1.

-

Jlisnilerereatedone

ams

mer)le 2Ilidspen-ricessilts

ream:t No.3) andlatoryat thepecial

Curve was fittedby eye.

5 IO I5 20Percentage of bottom sample by volume

passing through 0.883-mm. sieve

Figure 6.0. Relationship observed between coefficientof permeability and the fraction of the total volume ofstream bottom materials passing through an 0.833-mmsieve. (Curve fitted by eye.)

According to unpublished data on weir countsand visual estimates, the number of adultpink salmon spawning in Anan Creek on oc-casion has exceeded 1 million, and in mostyears exceeds 100,000. Five random sampleswere obtained from a large spawning rifflein Anan Creek about 3 miles upstream fromthe estuary. Even with this small number ofsamples, it was apparent that virtually nosilt and little fine sand were present andthat bottom materials were highly permeable.

Estimated escapements of pink salmon havebeen lower in Maybeso Creek than in any ofthe other streams sampled. A few hundred pinksalmon spawn some years in the intertidalzone of Maybeso Creek, where 20 percent ofthe total volume of solids present in an inter-tidal riffle were found to pass through an0.833-mm. sieve, indicating extremely lowpermeability.

Escapement estimates indicated that in themost years Twelvernile Creek was considered

Relative escapement

Very high

Medium to high

Medium to high

Harr isupstreamintertidol

II

Foir to medium

Twelvemileupstreamintertidal Low to fair

Movbesointertidal Low

I I I 15 10 15 2 0

Percentage of solids passing through 0.833 -mm. sieve

Figure ‘I.--Percentage of the total volume of bottommaterials in six Southeastern Alaska pink salmonstreams passing an 0.833-mm sieve. Samples werecollected in summer 1959. The streams have beenranked in accordance with approximate levels of pinksalmon escapements weighted by available spawningarea.

to be a marginal spawning stream, although itsupported larger runs of pink salmon thanMaybeso Creek. Although Twelvemile Creekcomprises several miles of spawning ground,pink salmon escapements rarely exceed a fewthousand adults. Size composition of bottomsamples shows that intertidal and upstreamspawning grounds in Twelvemile Creek con-tain large volumes of fine sands and silts andhave low permeability.

Estimates of pink salmon escapements intoHarris River, Indian Creek, and Old Tom Creekhave varied greatly, but sizable runs havebeen recorded for each of these streams insome years. Runs into Harris River and OldTom Creek commonly include tens of thousandsof adult pink salmon and may exceed 50,000on occasion. Both streams have extensivespawning grounds, with Harris River beinglarger. Spawning grounds in Indian Creek arelimited by comparison, but escapements ofthousands of adult pink salmon commonly

.7 ..

- . . . * .W”. .- . . ..I_ I

Table 2.-- Mean percent by volume of solids retained by sieves (openings in mm.) and settlingfrom suspension in 10 minutes, (All rocks larger than 4 inches minimum dimension excluded.)

Stream and8 ampling

areaSamples

Percent of total volume of solids retained by sieve total volumeAres with opening of -- of solids set-

sampled 26.26 13.33 6.68 3.33 1.65 0.833 0.417 0.208 0,104 tling fromsuspension

Anan CreekNumber Square feet

Upstream 5 ‘25,000 41.5 10.4 10.3 10.1 9 . 6 10.6 3 . 9 1 . 3 0 . 3 0.2

Indian CreekIntertidal 50 36,560 35.2 15.3 12.9 11.3 7 . 3 8 . 6 4 . 9 1 . 3 0 . 4 2.7

Old Tom CreekIntertidal 15 ‘36,000 26.5 17.8 15.7 12.4 7 . 4 8 . 6 6 . 3 2 . 4 0 . 7 2.2

Harris RiverUpstream 25 83,750 24.0 16.2 14.1 11.8 8 . 4 11.8 9 . 2 2 . 4 0 . 4 1 . 9

Harris RiverIntertidal 37 62,640 25.1 14.6 13.2 11.0 7 . 9 11.5 10.5 2 . 7 0 . 4 3.1

Twelvemile CreekIntertidal 50 60,080 2 1 . 2 15.1 13.9 12.5 8 . 8 10.7 8.4 2 . 5 0 . 9 5 . 9

Twelvemile CreekUpstream 50 66,400 19.7 13.9 13.0 12.9 9 . 6 11.6 9 . 5 3 . 6 1 . 2 5 . 0

Maybeso Creek 1Intertidal 16 15,000 19.4 11.3 12.6 13.4 11.0 12.3 10.1 4 . 7 1 . 4 3.8

1 Area estimated.

-.-*-t.

!1i

occur nevertheless, with peak escapementsapproaching 10,000. Harris River and Indianand Old Tom Creeks may be somewhat repre-sentative of numerous Southeastern Alaskapink salmon streams which, when taken asa whole, contribute significantly to the totalcatch. Although the volume of fine sands andsilts in these streams varied, it was apparentthat each contained considerably larger vol-umes of fine materials and was less per-meable than the area of Anan Creek sampled.

SILT REMOVAL BY SPAWNERSIn view of the nearly conclusive evidence

that silt is harmful to salmon eggs and larvae(see the review by Cordone and Kelley, 1961),it is likely that redd digging benefits progenythrough removal of fine particles and organicdetritus from spawning beds. Changes occur-ring in the composition of bottom materialsduring the spawning period were studied.

Volumes RemovedThe intertidal Harris River sampling area

was sampled randomly before and after spawn-ing in 1959 and 1960. The number of femalesspawning between the sampling dates wasestimated to be 2.4 per 100 square feet in1959 and 4.6 per 100 square feet in 1960.It was found that the percentage of solidspassing through an 0.833.mm. sieve decreasedsignificantly over the spawning period in bothyears (table 3). The greatest decrease occurredamong settleable solids passing through a0.104-mm. sieve and classified as ‘*pan” inthese studies. Other studies suggested that aconsiderable portion of fine particles removedby spawners consisted of light organic ma-terial.

Organic ContentThe organic content of pan materials was

determined for samples collected beforespawning in the summer of 1958. Nine samplesfrom spawning riffles in Harris River and In-dian and Twelvemile Creeks were dried andsorted. Materials passing through a 0.104-mm.sieve (pan) were divided into two groups beforebeing weighed--those retained by a 0.074-mm.sieve and those passing through a 0.074-mm.

sieve. The organic fraction of each group wasburned off, and the samples reweighed. Solidsretained by the 0.074-mm. sieve had an averageorganic content of 3.9 percent (range, 1.7 -7.2 percent); whereas the average organiccontent of solids passing a 0.074.mm. sievewas 12.4 percent (range, 7.9-19.0 percent).

Specific gravity of materials retained byand passing through the 0.074-mm. sieve wasalso determined. Mean values of four sampleswere 2.68 g./cc. for materials retained by the0.074-mm. sieve and 2.38 g./cc for materialspassing through the 0.074.mm. sieve. Thusit appeared that the highest organic fractionwas in the finest size groups, which were alsomost susceptible to complete removal fromspawning beds.

SPAWNING BED SILTATION

During our studies, we observed one instancewhere siltation of spawning beds occurred inassociation with logging. We also observedthat flooding removed most if not all of thesilt accumulating from logging.

Silt Pollution from Logging

Land areas contiguous to the upstream sam-pling area of Harris River were logged inlate summer and autumn 1959. From personalobservation, it was apparent that tributarystreams draining the logged areas carriedunusually heavy silt loads during freshets.Water samples collected from three tributarystreams during freshets contained, on theaverage, two to four times more inorganicsediment after logging then before. The av-erage amount of organic sediment in samplescollected after logging was only fractionallymore than the amount in samples collectedbefore logging. Because of the high variabilityobserved in the amount of sediment carriedby the tributary streams and the small num-bers of samples collected, it is not possibleto show conclusively that the content of in-organic sediment increased because of logging.However, the data, which are summarized intable 4, lend support to this conclusion.

9

0

.

Table 3, -- Mean percent by volume of fine sands and silts in samples from intertidal spawningbed of Harris River before and after spawning of pink salmon in 1959 and 1960

Year andtime sample s

collected

1959

Femalesspawningper 100

sq. ft.

Number

Total volumeof solids retained by

sieve with opening of--Samples 0 ,417 0.208 0.104 Pan

m m . m m . mm.

‘Number Percent Percent Percent Percent

Solids passingthrough

0 .833 .mm.sieve

Percent

37 10.5 2.7 0.4 3.1 16.7

17 9.4 2.1 0.4 1.3 13.21.1 0.6 0.0 1.81 3.52

32 7.9’ 1.7 0.9 4.3 14.8

23 8.0 1.8 0.2 1.5 11.50.1 0.1 0.7\1 2.81 3.32

- - -1 Difference significant at 0.5.percent level.

2 Difference significant at 5, O-percent level.

ti

I Table 4, --Suspended sediment content of three small tributaries ofHarris River before logging (1956) and after (1959)

Stage of HourWater Sediment contentI logging of

a n d d a t e daydepth Organic . Inorganic

i

/ TRIBUTARY AI

Before loggingAugust 30September 25September 30November 20

MeanAfter logging

September 11September 25October 11

Mean

1600

12301245

120011451345

Feet Mg. / l Mg/l1.2 3 2 1

0 . 4 53 100 . 5 56 1 5

1.1 9 1438 10

0.6 62 20

0.8 56 2 3

0.8 59 85

59 43

TRIBUTARY B

Before loggingAugust 30 1432 0.6 69 16September 25 1600 0 . 5 60 1 6September 3 0 0.6 52 62November 20 1307 0 . 9 39 1 5

Mean 55 27After logging

September 25 1200 0.5 35 46September 25 1200 0.4 6 1 49October 11 1130 0 . 7 53 49October 11 1350 0.5 85 155October 16 1645 0.5 96 20October 17 0830 0.9 113 73

Mean 74 - 6 5

TRIBUTARY C

Before logging

-August 30 1400 82 3

I September 25 1614 0 . 4 35 1September 30 1315 0 . 7 6 1 40/November 20 1350 1.0 23 14

Mean 50 14: After logging

October 11 0800 0.6 30 20October 11 1030 0.8 44 70October 11 1100 0.8 84 181October 13 1100 1.3 68 51October 16 1200 0.6 48 1 3,

I October 16 1700 0.9 82 0Mean 57 56

-.-- .

t

. .

Silt content of spawning beds.--Bottom sam-ples obtained from upstream in Harris Riverin August and October 1959 revealed that thevolume of silts and fine sands in spawningbeds increased significantly during periodswhen tributary streams appeared to carryincreased amounts of inorganic sediments.Similar increases did not occur in intertidalHarris River, suggesting that settleable solidswere not transported downstream in signifi-cant amounts, or in Indian Creek, which wasunlogged. Instead, the volume of fine particlesin intertidal Harris River and Indian Creekdeclined significantly between late summerand midautumn, possibly because of spawn-ing activity. Mean percentages of materialspassing through an 0.833.mm. sieve were:

Harris River (upstream):August 16, 1959--13.9 percentOctober 27, 1959--17.4 percent

Harris River (intertidal):August 13, 1959.-16.7 percentOctober 20, 1959.---13.8 percent

Indian Creek (intertidal):August 11, 1959.-9.3 percentNovember 10, 1959.-7.4 percent

For each area, differences between samplingdates were tested with a t-test and were foundto be significant at the 5-percent level.

The decrease in percentage of materialspassing through an 0.833.mm. sieve in inter-tidal Harris River was thought to be causedby redd digging (refer to table 3). Because thedensity of female spawners in Indian Creek wasnearly twice that observed in intertidal HarrisRiver in 1959 (4.3 versus 2.4 per 100 squarefeet), a decrease in the volume of fine particleswas to be expected here, too. By comparison,spawning females were scarce in upstreamHarris River in 1959 (0.6 female per 100square feet), and redd digging must have re-duced the silt content of upstream HarrisRiver only slightly. Hence, logging silt prob-ably did not replace a deficit created by redddigging, but simply added an amount additionalto what was already present in August 1959.

Effect of silt on permeability.--Accordingto the curve relating content of fine particles

to permeability (fig. 6), permeability of spawn-ing beds in upstream Harris River was re-duced considerably by settleable solids addedin autumn 1959. This was confirmed by alaboratory study.

After size composition was determined,gravel samples from intertidal Harris Riverand Indian Creek were placed inconstant-headpermeameters (fig. 5). A mixture of fineparticles (5 parts retained by 0.208-mm.sieve; 3, by 0.104-mm. sieve; and 12, by pan)was added to each permeameter. The amountof fine materials added varied from 2.4 to2.8 percent of the total volume of each sampletested. (The estimated increase in volume offine particles in upstream Harris River duringautumn 1959 was 3.5 percent.) Permeabilityof each sample before and after addition offine particles is given in table 5. Addition offine particles reduced permeability an averageof 2.5 times (range was 1.4 to 4.3 times).

Silt Removal by Flooding

The deposition of fine materials in the up-stream Harris River sampling area was foundto be temporary. Flooding occurred in theperiods November 5-7 (5.32 inches of rainin 72 hours) and December 5-7 (7.02 inchesof rain in 72 hours). ’ In samples of bottommaterials obtained from upstream HarrisRiver after flooding (February 23, 1960).the content of fine materials had returnedto the level observed in summer 1959 beforelogging.

Observations on individual size groups pass-ing through an 0.833.mm. sieve revealed thatsizes retained by 0.208-mm. and 0.104-mm.sieves and passing through a 0.104.mm. sieve(pan) were most affected by logging and flood-ing. Mean percentages of the total volume ofbottom material in upstream Harris River inthese four size groupings plus those retainedby a 0.417-mm. sieve show changes involumesof fine particles occurring in association withsiltation and flooding (table 6).

4Rainfall was recorded at Hollis by the U.S. ForestService.

1 2

.

l-3-

:da

d,2radnen.n)nttobleofw3ityofofl3e

‘P-lnd:heainles3mrisIO).ledxe

SS-hatLm.eveDd-? ofr innednesvith

rest

Table 5 .--Decrease in permeability of bottom materials from salmon spawning beds withaddition of fine particles 1

?,

i [Permeability readings corrected for standard viscosity of 1.0 centipoise at 680 F.]I

I

iNo fine particles added

percent passingIi

through Permeability

i 0.833-mm s i e v e

Fine particlPercent passing

through0.833-mm. sieve

I -I I 6.1!I 4.5

I14.7

t 10.3

I10.9

I9.7

12.7Ii 5 . 6

Cm. /min.270 8.8

510 7.3

29 17.1177 12.8

5 8 13 .4

163 12 .443 15.3

313 8.1

es a d d e d I

Permeability

Cm/min.80

36214

994057

1 0173

1Fine particles added consisted of materials retained by a 0,208.mm,

(5 parts) and a 0.104-mm (3 parts) sieve and passing through a 0.104-mm.sieve (12 parts),

Table 6.--Mean percentage of total volume of bottom materials in five size classes of fineparticles from upstream Harris River before siltation from logging, after siltation, and

after siltation and flooding

Condition of stream

Before siltation(August 10, 1959)

After siltation(October 27, 1959) 1

After siltation andflooding

(February 23, 1960) 2

Passing through Retained by Retained by Retained by Passing through0.833-mm. 0.417.mm. 0.208-mm. 0.104-mm. 0.104-mm.

seive sieve sieve sieve sieve

Percent

13.9

17.4

14.0

Percent Percent

9.2 2.4

9.7 4 .2

Percent

0.4

0 .6

Percent

1.9

2.9

9 . 2 2.7 0.3 1.8

Size group

, *Harris River was observed to carry heavy silt loads on October 12 and 17, 1959.2 Flooding occurred on about November 6 and December 6, 1959.

I,

P;

.

.

‘.

.

.

L

.

.

L

‘.

.,

. .

.

‘.

.

.

Differences between mean value6 Of August16 and October 27 and of October 27 and Feb-ruary 23 were evaluated with a t-test. Exceptfor the fraction retained by the 0.417-mm.sieve, all differences were significant at thel&percent level at least.

SUMMARY

Permeability of bottom materials is prob-ably an important factor limiting fry productionin spawning beds. It has been shown howpermeability is related to size compositionof bottom materials, and it is possible tocompare relative permeability of spawningbeds by determining average size composi-tion. Hence, a simple and direct method ofmeasuring physical properties of spawningbeds pertaining directly to their capacity toproduce fry has been provided through develop-ing techniques for measuring size composi -tion of bottom materials.

Pronounced differences in size compositionof bottom materials were observed amongseveral pink salmon spawning streams. Themore productive spawning streams had thehighest permeability coefficients.

The volume of fine materials within par-ticular spawning beds changes with time.A reduction in volumes is caused primarilyby forces that produce gravel movement (e.g.,redd digging and flooding). Silts and fine sandsaccruing to spawning beds and reducing theirpermeability are transported in part fromwatersheds into spawning streams. Loggingincreases the amount of transported materials.

LITERATURE CITED

ANDREW, F. J., and G. H. GEEN.1960. Sockeye and pink salmon production

in relation to proposed dams in theFraser River system. International Pa-cific Salmon Fisheries Commission,Bulletin No. 11, 259 p.

BURNER, CLIFFORD J.1951. Characteristics of spawning nests of

Columbia River salmon. U.S. Fish andWildlife Service, Fishery Bulletin 61,vol. 52, p. 97-110.

.

CHAMBERS, JOHN S., GEORGE H. ALLEN,and R[ICHARD] T. PRESSEY.

1955. Research relating to study of spawn-ing grounds in natural areas. A n n u a lreport (June 22, 1954 to June 22, 1955),Washington Department of Fisheries toU.S. Army Corps o f Eng ineers .[Processed.]

CHILDS, E. C., N. COLLIS-GEORGE, andJ. W. HOLMES.

1957. Permeability measurements in thefield as an assessment of anisotropyand structure development. Journal ofSoil Science, vol. 8, no. 1, p. 27-41.

CORDONE, ALMO J., and DON W. KELLEY.1961. The influences of inorganic sediment

on the aquatic life of streams. Cali-fornia Fish and Game, vol. 47, no. 2,p. 189-228.

FAIR, GORDON M., and L[ORANUS] P. HATCH.1933. Fundamental factors go v e r n i n g

streamline flow of water through sand.Journal of the American Water WorksAssociation, vol. 25, no. 11, p. 1551-1565.

FRANZINI, JOSEPH B.1951. Porosity factor for case of laminar

flow through granular media. Transac-tions American Geophysical Union, vol.32, no. 3, p. 443-446.

FREVERT, RICHARD K., and DON KIRKHAM.1948. A field method for measuring the

permeability of soil below a water table.National Research Council, Proceedingsof the Highway Research Board, vol. 28,p. 433-442.

HARRISON, C. W.1923. Planting eyed salmon and trout eggs.

Transactions of the American FisheriesSociety, vol. 53, p. 191-199.

HATCH, RICHARD W,1957. Success of natural spawning of rain-

bow trout in the Finger Lakes regionof New York. New York Fish and GameJournal, vol. 4, no. 1, p. 69-87,

1 4

ALEN;‘l

pawn-1nnual1955),:ies toe e r s .

:,, a n d

in theotropynal of27-41,

,LEY.iiment

C ali-no. 2,

1TCH.-ningI sand.Works1551-

.minaransac-n, vol.

:HAM.lg t h e* table.edings‘01.28,

: eggs.heries

! rain-regionGame

KROCHIN, SVIATOSLAV.1955. Flow of water through lead shot

mixtures influence of uniformity onfraction factor. M.S. thesis, Universityof Minnesota, Minneapolis, 84 p.

LUCAS, K. C.1960. The Robertson Creek spawning chan-

nel. Canadian Fish Culturist Issue 27,p. 3-23. .

MARTIN, JOHN WILSON (editor).1959. Stream catalog of eastern section of

Ketchikan management district of South-eastern Alaska. U.S. Fish and WildlifeService, Special Scientific Report-Fisheries No. 305, 379 p.

POLLARD, R. A.1955. Measuring seepage through salmon

spawning gravel. Journal of the Fish-eries Research Board of Canada, vol.12, no. 5, p. 706741.

SHAPOVALOV, LEO.1937. Experiments in hatching steelhead

eggs in gravel. California Fish andGame, vol. 23, no. 3, p. 208-214.

SHAW, PAUL., and JOHN A. MAGA.1943. The effect of mining silt on yield

of fry from salmon spawning beds.California Fish and Game, vol. 29,no. 1, p. 29-41.

SHELTON, JACK M.

1955. The hatching of chinook salmon eggsunder simulated stream conditions. U.S.Fish and Wildlife Service, ProgressiveFish-Culturist, vol. 17, no. 1, p. 20-35.

TERHUNE, L. D. B,1958. The Mark VI groundwater standpipe

for measuring seepage through salmonspawning gravel. Journal of the Fish-eries Research Board of Canada, vol.15, no. 5, p. 1027-1063.

WICKETT, W. P[ERCY].1958. Review of certain environmental fac-

tors affecting the production of pinkand chum salmon. Journal of the Fish-eries Research Board of Canada, vol.15, no. 5, p. 1103-1126.

MS #1274