<br />I'
<br />J'
<br />~
<br />
<br />}
<br />1.
<br />
<br />~
<br />
<br />j
<br />
<br />~,
<br />,
<br />{
<br />f
<br />l'"
<br />t
<br />~
<br />~
<br />"
<br />i,
<br />f
<br />~.
<br />~~.
<br />~
<br />,
<br />~
<br />: ~
<br />
<br />L
<br />i'i'I'
<br />" ,
<br />
<br />ii:
<br />,~
<br />'~ .
<br />d1f
<br />os: ...
<br />~' f-,
<br />
<br />~
<br />~
<br />'.('
<br />;~ .'
<br />
<br />
<br />
<br />.
<br />Scour and fill in steep, sand-bed ephemeral streams
<br />
<br />Ic~D
<br />JUl 22 1997
<br />
<br />Colorado Water
<br />8 I _&.:_- 0__.....
<br />
<br />MICHAEL G. FOLEY Department of Geology, University of Missouri-Columbia, Columbia, Missouri 65201
<br />
<br />ABSTRACT
<br />
<br />The traditional idea that entire long reaches of alluvial stream
<br />channels in semiarid regions are scoured at high flood discharges
<br />and subsequendy filled in the waning flood phaseJmean-lxd scour
<br />and fill) can be challenged. The alternative concepc chac mean-bed
<br />elevacion varies buc little during a flood and chac both scour and fill
<br />occur concurrently at different migrating loci within a reach (local
<br />scouc and fill) is also consistent wich published field data. Field and
<br />laboratory" investigations reported herein suggest mat mean-bed
<br />scour and fill in a straight uniform channel is minor compared to
<br />local scour and fill caused by bedform migration and, furthermore,
<br />that maximum local scour and fill may occur during the waning
<br />flood phase in some instances.
<br />The field experiment. utilizing a rectilinear array of maximum-
<br />scour indicators (scour-cords). produced data for concoured plots
<br />of maximum scour and fill in an ephemeral stream bed during two
<br />floods. In the first flood, 24 em of scour and fill was measured for a
<br />bankfull flow depch of 23 em. In che s<cond, maximum scour and
<br />fill was at lease 66 cm for a bankfull flow depch of 34 cm. Escimaces
<br />of antidune amplitudes for che cwo floods, based on theocetical
<br />models and laboratory and field obs<cvarions, ace 27 to 61 em and
<br />44 to 92 em, respectively. This indicates that aU scour and fill mea~
<br />sured by the scour-cord array coUld have been caused by antidune
<br />migration.
<br />Laboratory experimems were conducted in an 18-m long non~
<br />recirculating flume with automated controls for rates of sediment
<br />and water input.. A series of experiments in a 26.7-cm-wide sand-
<br />bed channel wich rigid walls, at grade for a simulated flood pat-
<br />terned after those typical of ephemeral streams, showed that
<br />mean-bed scour and fill was less chan 3% of local <couc and filL For
<br />thc:se experiments, mean sand size was 0.3 mm, channel slope was
<br />0.009, maximum water depth was 40 mm, maximum local scour
<br />and fill waiU mm; and maximum mean-bed scour and fill was 0.6
<br />mm. Maximum mean-bed elevation variation was thus only two
<br />sand-grain diameters. Maximum local scour and fill took place
<br />near che end of the simulated floods. when bed-form amplitudes
<br />were me greatest'".
<br />Ancidune. amplitudes calculated for flows in che Arroyo de los,
<br />Frijoles) New Mexico, are larger than published values of scour and
<br />fill for unit discharges greater chan 0.5 m'/m . s (5 dslftl_ Below chac
<br />chreshold discharge, che antidunes thac formed at maximum flood
<br />discharge are smaller than the dunes that probably form during the
<br />waning flood, and calculated antidune amplitudes are less than re-
<br />poered'scour and fill. ,::r:.l"",'n: .1,0 >l '; T. ~ _.'1 , ,
<br />
<br />.7 '~:_i I; I );<:f: l:!~: ,.l., ':, :.;~ ~ i~\: J ~l>\~", ;,':.'~~.~' ~;~\~~": :,'.:-,~,: ;':;~,~:,'.: :""~.:,,,:,,_:
<br />~qDUCTIONi. L:,',.',Jd -1.W!" .~_. ."~I &. ...<~".J. u'-"" .' i '~_',
<br />:;;.. '. >!~;:"'~::;". :. I)'}iJ-m;' ;ri "dl1ir!:':fP:;'.,~~ih .;,; ,.:;~J i1j~:!(:,,;,"~o::~ ~t..;'!;ii'1
<br />. -.Steep;'sand-bed screams 'are 'subject- to high~v~lodty.fiOOd t10~.
<br />Large aInounts of suspended debris in these flows make-direct bed
<br />obserVations' impossible, and high velocities make', soundings
<br />difticulr and eXpensive. Thus, bed bchavioc during floods is poody
<br />Wldecs;ood.:.Soundings from '.bridges or. gaging station cable-
<br />
<br />crossings (for example, Pierce, 1916; Leopold and others, 1964)
<br />indicate that stream beds at grade are lowered by scour during
<br />flood-crest passage and raised by deposition to approximately their
<br />former elevation on the waning flood - a behavior known as scour
<br />and filL In discussions [Q follow, reference is made to "mean-bed
<br />scour and fill" and "lohl scouc and filL" "Mean-bed scour" is
<br />defined as scour occurring simultaneously OVer a stream-segment
<br />length comparable [Q the flood-crest length; the subsequent
<br />"mean-bed filling" is likewise defined. "Local scour and fill" is lim-
<br />ited to a reach whose length is comparable to or less chan stream
<br />width. Local scour and fill is related to flow disturbances occurring
<br />in narrow gorges or near bridge abutments and other engineering
<br />structUres, or to "bed-form scour and fill" related to bed-form de-
<br />velopment and migration. Local scour and fill does not necessarily
<br />occur simultaneously over the entire: transverse seerion of a stream
<br />channe~ and it may Occur several rimes at the same location during
<br />a single flood, "Total scouc and fill" is the combined effece of local
<br />and mean-bed Scouc and fill. "Scouc and fill" will be used inter-
<br />changeably wich "bed reworking" whece che specific nature of the
<br />process is unknown.
<br />The usual interpretation of gaging station measurements has
<br />been that the entire flooded stream reach experiences mean-bed
<br />scour and fill comparable to that in the measured cross sections.
<br />However, Mitchelllin Leopold and Maddock, 1952) indicated chac
<br />measurements at gaging stations may not be representative of the:
<br />whole flooded reach, Lane and Borland (1954), obsecving chat gag-
<br />ing stations are typically located in narrows, hypothesized that
<br />deposition occurs where the'stream widens downstream. They also
<br />proposed that restoration of the preflood c:levation of the bottom
<br />involves deposition in scoured seroons of material removed from
<br />such accumulation sections. From poine measurc:ments along cx-
<br />tensive reaches of both ephemeral and perennial streams, Emmett
<br />and Leopold (1963) found (1) thac SCOUt of che preflood bedoccucs
<br />at some time during flood passage at all sections of these {eaches
<br />and (2) chat in the two perennial scream; chey studied, chis scour
<br />occurred simultaneously over a long teach.
<br />These field observations are generally interpreted to indicate that
<br />scour occurs continuously on all or part of the stream bed until that
<br />process is replaced by deposition, when in fact scour and fiH may
<br />alternate locally several times during a flood. with only a small per.
<br />centage of the emire reach experiencing scour or fill at anyone time
<br />(CulbectSon and Dawdy, 1964). This implies chat the scour-and.fill
<br />process may be related to movement of bed forms such as dunes
<br />during flood passage. Colby (1964) concluded chac the maximum
<br />scour measured at a point is simply the elevation of the lowest in-
<br />terdune "trough [0 pass during a flood; he: also 'concluded .that
<br />mean-bed scour and fill occucs only in channel constrictions. He
<br />reache&. this conclusion ~by . sedimenr~transport 1.analysis of a
<br />hypotfi;;tical. flood in a unifonn sand-bed channel. However, his
<br />. analysis assumed a continuous stage: versus discharge relanon and
<br />slowly changing equilibrium flow condicions.
<br />Under SOme conditions of slope and bed-material size, sand-bed
<br />streams do not have continuous stage versus discharge relations.
<br />
<br />.'..
<br />
<br />"
<br />
<br />Geologia! Soci~ry of America Bulletin. v. sf. p. 559-570, 106gs., 1 ubl~ April 1978, Doc. no,.~0408.
<br />~', i' ':', ,,<-,t, ".
<br />
<br />559
<br />
<br />'"',,
<br />
<br />'.'':' ',.
<br />
<br />';1,"
<br />
|