University of Nebraska - LincolnDigitalCommons@University of Nebraska - Lincoln

US Army Corps of Engineers U.S. Department of Defense

1991

Reservoir Sedimentation StudiesD C. BoudurantU.S. Army Corps of Engineers

R H. LiveseyU.S. Army Corps of Engineers

Follow this and additional works at: http://digitalcommons.unl.edu/usarmyceomaha

Part of the Earth Sciences Commons, Environmental Sciences Commons, and the Oceanographyand Atmospheric Sciences and Meteorology Commons

This Article is brought to you for free and open access by the U.S. Department of Defense at DigitalCommons@University of Nebraska - Lincoln. It hasbeen accepted for inclusion in US Army Corps of Engineers by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.

Boudurant, D C. and Livesey, R H., "Reservoir Sedimentation Studies" (1991). US Army Corps of Engineers. 158.http://digitalcommons.unl.edu/usarmyceomaha/158

Reservoir Sedimentation Studies

D. C. BONDURANT AND R. H. LIVESEY

U.S. Army Corps of Engineers, Omaha, Nebraska 68101

It must be recognized that, when a reservoir isavailable for study is adequate, we establish a sta-

constructed on a stream, all or most of the sedi-tion at that location.

ment transported into the reservoir by the stream

Finally, we make an extensive ground recon-

will be deposited there. With few exceptions,naissance of the drainage area above the site. In-

there is no practical method for reducing orformation from this reconnaissance is correlated

eliminating the sediment inflow; thus it iswith all available data from stations measuring

necessary to anticipate and to provide for thesuspended sediment discharge from comparable

resulting problems. The major problem is, of

areas and is integrated to derive an estimate of

course, the accumulative loss of storage capacity.the average annual sediment inflow to be an-

Other items of importance include the distribu-ticipated. It is necessary to add to this estimate

tion of the sediment with respect to varioussome quantity to account for sediments moving

storage increments, the effect on the chemical oralong the stream bed and not measured in

physical quality of the water, ecological effects,suspended load sampling. In streams having a

and the possible degradation of the channel

coarse gravel bed, this load can be computed withdownstream. reasonable adequacy by bed load formulas;

The prediction of the quantity of sediment that

however, in sand bed streams it is believed that

will be transported into the reservoir is largely aan estimate based on judgment is equally or

matter of experienced judgment. During the pastperhaps more accurate. In small reservoirs the es-

20 years, there have been many stations es-timate may be reduced to account for sediment

tablished for the measurement of suspended sedi-that might be transported through the pool, but,

ment discharge, but the cost of operating thesewhere the drainage area contributing to the pro-

stations is such that they are restricted in number

ject is greater than about 250 km', the reservoir

and location to a relatively few index areas or towill normally be large enough to retain all the in-

specific locations where they may be operated for

flowing sediment. If the project is constructed un-

only a few years. Because of the normal

der the auspices of the federal government,

variations in the hydrologic cycle a station mustsufficient storage is provided to retain the an-

usually have a record of 10-30 years (dependingticipated sediment load for a 100-year period

on the physical and hydrologic characteristics of

without encroaching on the primary project pur-

the area) in order to provide a dependableposes.

average. Since it is seldom that the need for sedi-In some instances it is possible that the storage

ment data at a specific location can be predicted

required for sediment might be reduced by up-

that far in advance, it is also seldom that ade-stream control measures. It has been demon-quate records are available. strated, for example, that the sediment contribu-

In the Missouri River division of the U.S.tion from very small drainage areas (1-5 km') can

Army Corps of Engineers we maintain certain in-be reduced by about 85% by intensive soil conserva-

dex stations continuously. In regions where thetion procedures. In another instance, sediment dis-

sediment discharge is reasonably consistent wecharge reductions due to improved land manage-

operate stations at one location for 5-10 years;ment and conservation on a group of drainages

then, providing for an overlap of 1 or 2 years, wevarying in size from several hundred to 5000 km'

move to other locations to establish index valueswere indicated to be 10-35%. This latter study,

for the region. As soon as the need for data at ahowever, covered a period of only 3 years under the

specific location is known and if the timeimproved regimen and could well have been in-364

Geophysical Monograph Series Man-made Lakes: Their Problems and Environmental Effects Vol. 17

Copyright American Geophysical Union

BONDURANT AND LIVESEY

365

fluenced by favorable hydrologic circumstances.In general, it is usually found that those por-

tions of the drainage area contributing thegreatest proportion of the sediment are the areaswhere conservation measures are the leastjustified economically for improvement of theland. As an example of this condition, <15% of thestructures proposed in a comprehensive conserva-tion plan for one area have proved to be justifiedeconomically. In addition, the installation andcontinued maintenance of conservation measuresdepend wholly on the cooperation of the land-owners. When it is further considered that therewill already be a large quantity of sedimentavailable for erosion in the contributing streamsystem, the sediment discharge reduction availablefrom upstream conservation will probably be lessthan the errors inherent in the basic evaluation ofthe sediment storage requirements. Although con-servation benefits to the land itself cannot be deniedand flood flow reductions due to upstream struc-tures will undoubtedly be beneficial to a reservoirproject, it is unwise, except in rare circumstances,to reduce sediment storage requirements in an-ticipation of benefits from upstream controls.

In a reservoir where the entire operatingstorage is reserved for one purpose, the locationof the sediment deposited in a reservoir might notbe important. In multiple-purpose reservoirs,however, where the storage in varying zones ofelevations is reserved for individual purposessuch as power production, irrigation, recreation,and flood control, it is desirable to know just howmuch storage will be lost in each zone owing tosedimentation and what the backwater effect onupstream facilities might be. This effect will de-pend on several factors such as the characteristicsof the sediment, the chemical character of thewater, the size and shape of the reservoir, theoriginal stream and valley slopes, and the reser-voir operation. As soon as the stream enters thebackwater reach above the then existing pool, thelargest sediments in transport begin to deposit inthe channel. The deposition occurs progressivelyuntil, some distance within the pool, all the sand-sized materials are deposited. This progressioncontinues with the silt and, finally, the clay-sizedsediments.

If the pool is relatively narrow (not more than3 or 4 times the normal width of the streamchannel), the sand materials will be depositedacross the entire width of the pool, and, as theprocess continues, these deposits will form a delta

that gradually extends downstream. Depositionof the silt-sized sediments occurs more or less im-mediately downstream from the sands, and thedeposition of clays occurs either a short distancethereafter or perhaps much farther downstream,depending on the chemical characteristics of thesediment and water.

If the stream enters a wide pool, deposition ofthe sand sediment tends to progress as a finger,and a submerged channel is formed into the pool.A finite flow will continue along this submergedchannel and will result in a reverse circulation inwhich deposition of a large portion of the silt andclay sediments will occur in that part of the pooladjacent to the sand finger. As the process con-tinues, the sand finger extends to the water sur-face and forms a surface channel extending intothe pool. This channel, possibly as a result ofCoriolis forces, tends to follow the western orsouthern boundary of the pool in the northernhemisphere. Vegetation growing on the bankstends to hold this channel in place; however, ahigh flow will ultimately breach the channel andwill start a new finger. The result is a mixture ofclay, silt, and sand intrusions forming a swampy,vegetated area that may contain random pools ofopen water.

The character of the clay sediments in conjunc-tion with the chemistry of the water plays an im-portant role in the deposition of the claymaterials. The active clays, the montmorillonitegroup, may react with the dissolved salts in thewater in a manner such that the particles have amutual attraction and tend to form clumps offloccules that settle out of the water with relativerapidity. On the other hand, these clays may forma low-density, semifluid mass that may ac-cumulate downstream from the sand delta, maytravel through the reservoir as a density flow, ormay accumulate immediately at the head of thereservoir. Which of these phenomena will occuris governed by the activity and concentration ofthe clays, the character and concentration of thedissolved salts in the water, and the slope of theoriginal stream channel and valley. Sedimentdensity flows of any magnitude apparently do notoccur with slopes of <0.2 m/km.

The inactive clays, the kaolinite group, do notreact as readily as the active clays with the salts inthe water, and the particles tend to be mutuallyrepulsive unless they are present in sufficient con-centration for mass attraction forces to be domi-nant. The finer particles may remain in suspen-

Geophysical Monograph Series Man-made Lakes: Their Problems and Environmental Effects Vol. 17

Copyright American Geophysical Union

366

RESERVOIR SEDIMENTATION STUDIES

sion in the reservoir for days or even weeks andthus may maintain turbidity throughout the en-tire reservoir.

It is only in rare cases that sediment can beevacuated from a reservoir except by mechanicalmethods, and the use of these methods is seldomwithin economic reason. It is true that the debrisbasins used primarily in the western coastalregions of the United States to protect high-valueurban areas from sediment and debris aredeveloped to be cleaned by physical removal ofthe material, but these basins are a separate con-sideration and normally have no connection withstorage reservoirs. In a few instances whereopenings have been made in existing dams in anattempt to flush out sediment deposits, the onlyresult was the erosion of a deep, narrow channelto the outlet. A few small detention-type reser-voirs are known to be self-cleaning, but in theseinstances the width of the pool is not muchgreater than that of the original stream channel.

There are instances where the character of thesediment and the water complex is such that den-sity flows can be used to evacuate the reservoirsand maintain storage capacity. Insofar as isknown to the authors, these instances are all inAfrica, primarily in Algeria. Here, the sediment ispermitted to accumulate until it reaches apredetermined density; then outlets specificallyprovided for the purpose are opened, and thesediment is evacuated with a minimum waste ofwater. In another instance a subdam at the headof the pool is designed to accumulate water andcoarser sediments and then to release the entiremass through the pool. A detailed description ofthese operations is given in papers presented byH. Duquennois and J. Thevenin and reproducedin the minutes of an international colloquium ondams and reservoirs held at the University ofLiege, Liege, Belgium, in May 1959.

In recent years, problems of ecology, waterquality, and recreation have assumed major im-portance in the planning and operation of reser-voirs. In many instances the influence of sedimen-tation on these functions is not completelyknown; however, there are many ramificationsthat need to be analyzed. Sediments deposited inrelatively shallow areas tend to prevent thereproduction of many of the media on which fishfeed or tend to result in a variation of the foodpattern such that the growth of one fish species isinhibited and that of another is enhanced. Theformation of a delta may block the passage of fish

that must travel to the open river upstream forspawning, and at the same time the delta mayform a swampy area that provides a desirablehabitat for entirely different species of wildlife.Inactive or dispersive-type clays tend to remain insuspension for relatively long periods and spreadthroughout the reservoir to create a turbid condi-tion. These clays will adhere to the micro-organisms forming a part of the fish food cycle andwill carry these microorganisms to the bottom asthe particles settle; thus the food required for fishlife is destroyed.

Sedimentation also plays an important role inthe eutrophication processes of lakes and reser-voirs. Organic material transported into the pooldecomposes; during this process, available ox-ygen is used, and, at the same time, nutrients arereleased. These and other nutrients transportedby the sediments result in accelerated biologicalactivity and overproduction by both plants andanimals within the photosynthetic region. Theseplants and animals, in turn, die off and ac-cumulate at the bottom where they decomposeand start the cycle again. In shallow waters theexcessive growth of aquatic plants so generatedmay completely fill the reservoir; however, withinreas'onable limits it is beneficial to fish life.Dispersive clays that create a turbid condition inthe waters restrict the photosynthesis and thusrestrict the biological activity. Such a conditionmay be desirable for swimming, boating, andother water contact sports for which an accu-mulation of aquatic growth would be detrimental.

Since sediments provide large surface areasfor chemical action, they may contributesignificantly to the rapid degradation or detox-ification of pollutants. Pollutants attached tosediment particles are not dispersed or transportedas readily as dissolved pollutants. Large concen-trations may build up in the stream bed or reservoirdeposits. These pollutants might be removed fromthe water environment by burial in the deposits, orthey might accumulate and be released into solu-tion by a later significant change in the waterchemistry. One of the more beneficial aspects is theremoval of pesticides from solution by chemicalreaction with clay materials.

The physical effect of a reservoir on thedownstream channel is also an important con-sideration. After deposition of sediment in thereservoir the clear water released through thedam will have an unsatisfied transport capacitythat it will attempt to satisfy by eroding the bed

Geophysical Monograph Series Man-made Lakes: Their Problems and Environmental Effects Vol. 17

Copyright American Geophysical Union

BONDURANT AND LIVESEY

367

and banks of the stream. The total capacity fortransport will be substantially reduced, however.because of the elimination of flood flows by reser-voir operation. The degree of reduction will varyfrom 25% to as much as 75%, depending on therelative character of the controlled releases ver-sus the normal flows. It should be noted,however, that in some instances the magnitude ofbank caving may be severely increased if thereservoir operation results in sustained periods of

near bank-full flow. These sustained periods per-mit the banks to become saturated, and thus theyare more susceptible to caving during sustainedperiods than during short periods of high flows.The effects of the downstream erosion mayinclude degradation of the channel, possiblehead cutting in tributaries, damage to down-stream bridges or other facilities, and possiblydamage to outlet works or hydroelectric powerunits.

Geophysical Monograph Series Man-made Lakes: Their Problems and Environmental Effects Vol. 17

Copyright American Geophysical Union