University of South Carolina
Scholar Commons
Faculty Publications
Geography, Department of
1-1-1983
Variability of Sediment Removal in a Semi-Arid
Watershed
William L. Graf
University of South Carolina- Columbia, [email protected]
Follow this and additional works at: http://scholarcommons.sc.edu/geog_facpub
Part of the Geography Commons
Publication Info
Published in Water Resources Research, Volume 19, Issue 3, 1983, pages 643-652.
http://www.agu.org/journals/wr/
© 1983 by American Geophysical Union
This Article is brought to you for free and open access by the Geography, Department of at Scholar Commons. It has been accepted for inclusion in
Faculty Publications by an authorized administrator of Scholar Commons. For more information, please contact [email protected].
WATER RESOURCES
RESEARCH,
VOL. 19, NO. 3, PAGES 643-652, JUNE 1983
Variability of SedimentRemoval in a SemiaridWatershed
WILLIAM
L. GRAF
Departmentof Geography
and Centerfor Southwest
Studies,ArizonaState University,Tempe,Arizona85287
Field and documentarydata from Walnut Gulch Watershed,an instrumentedsemiariddrainagebasin
of approximately
150km2 (57mi2)in southeastern
Arizona,showthat 83% of thealluviumremovedfrom
the basinduringa 15-yearerosionepisodebeginningabout 1930wasexcavatedfrom the highest-order
stream.The amount of alluvium removedin the erosionepisodewould have beenequal to a coveringof
about4 cm (1.6in) overthe entirebasin.The rate of sedimentremovalduringthe erosionepisodewas 18
timesgreaterthan therate of presentchannelsedimenttransport.Productionof sedimentfrom slopesand
channelthroughputat presentratesare approximatelyequal,and refillingwill not occurunderpresent
conditions.The channelformsleft by the massiveevacuationof sedimentimposecontrolson the spatial
distributionof tractiveforceand total streampower that make renewedstorageof sedimentlikely in only a
few restrictedlocations. Modern instrumentedrecords of a decade or more provide an inadequate
perspectiveon long-termsedimentmovement.
INTRODUCTION
The geomorphologistand engineersharea commoninterest
in the transport and storage of sediment in semiarid watersheds.Using theoreticaldeductions,the engineerhasfrequently
attemptedto predict the behavior of sedimenttransport systems,while the geomorphologisthas frequentlyusedfield observationsto inductivelyarrive at explanationsfor the process
[Shen, 1979, p. 20/10]. The following paper representsan attempt to combine engineeringand geomorphologicalperspectiveson the sedimenttransportprocesses
in channelsof a
semiarid watershed, Walnut Gulch in southern Arizona. Sevday prowell doc
n
the watershed[e,g.,Renardand Laursen,1975]. The emphasis
inthe presentpaperis on basinwideconsiderations
of almosta
centuryof processoperation.
There are three basic researchquestionsregardingthe variation in sediment transportation in the Walnut Gulch Watershed. First, how has sedimenttransport varied through time?
Short-termstudieshaveprovidedestimatesof the magnitudeof
presentsedimenttransportprocesses
on slopesand in channels
whichmay be placedin a contextwhencomparedwith centurylong record. Second,what has been the spatial distribution of
sedimentremoval during the last erosionepisodewhen arroyo
developmentoccurred?Significantchannelerosion and concomitant sedimentremoval occurredin the watershed during
the major erosionepisodethat spannedmost of the southwestern United States [Cooke and Reeves,1976]. Third, what has
that erosion implied for continued sedimenttransport? The
channel morphology left by the catastrophicerosion episode
representsa geometry that controls presentlyoperating processes[Cooke andReeves,1976].
STUDY AREA
The Walnut
Gulch Watershed
is located in southern Arizona
on pediment gravels,limestone,and igneousoutcropsnear the
town of Tombstone(Figure 1). The drainage area above the
lowest U.S. Department of Agriculture measurementsite is
about 150km2 (57.7mi2).The rollingterrainof the basinhas
sandy and gravelly soils typical of the Basin and Range geCopyright1983by the AmericanGeophysicalUnion.
omorphic province [Hunt, 1974]. Geologic materials include
Precambrian volcanicsand tertiary alluvium [Gilluly, 1956;
Wilson, 1962]. Elevation rangesfrom about 1220 m (4000 ft) to
about 1890 m (6200 ft). The climate is semiarid with mean
annual precipitationof about 35.6 cm (14 in) and mean annual
temperature of about 17øC (63øF), with wide ranges for both
variables [Sellers and Hill, 1974]. Precipitation falls mostly as
rain from winter stormsor suddensummerthunderstorms,so
that the entrenchedchannelsin the basinare usuallydry. The
vegetationin the basinis grass,shrubs,and brush.
Human impacts in the basin include the townsite of Tomb-
stone and associatedpreciousmetal mines begun in 1877,
[Meyers,1956].The mineshaftsprovidedlittle surficialdisturbance,but the wastematerialsfrom the shaftsand from milling
operations,active until about 1930,impactedlimited areasof
the landscape.A railroad servingthe mining area and numerous cross-basinroads may have affected channel processes,
especiallythrough stabilizingthe channelgradient by hardening road/channelintersections.In 1953 the U.S. Department of
Agriculturebeganan intensiveinstrumentationprogram in the
basin which
included
the installation
of numerous
concrete
flumes in the channel system[see Ferreira, 1979]. The flumes
resultedin further stabilizationof establishedstreamgradients.
Cattle grazing has affectedvegetationcover and related hydro1ogic/geomorphicprocessessince Spanish incursions from
Mexico more than two centuriesago.
Stream channels of Walnut Gulch watershed were mostly
narrow and shallow in the late 1800's and meandered
across the
upper surfacesof alluvial fills. Plat maps drawn in 1881 by the
General Land Office Survey show cienegas(wide grassymea-j
dows) in a number of flow areas, especiallyat junctions of
valleysin the center of the basin. In the lower reachesof the
trunk stream, the channel was wide, shallow, and sandy. Beginning in about 1930, residentsof the area report that the
channelsof the entire drainage basin were entrenchedand the
cienegaswere destroyed. Massive amounts of sediment were
evacuatedfrom the basin, as occurredin many streamsof the
semiarid and arid southwestern United States [Cooke and
Reeves,1976]. The entrenchedchannelsremainedin 1981,with
small amountsof refilling,as observedelsewherein the American southwest[Emmett, 1974; Leopold,1976].
METHODS
Analysis of sediment removal during the erosion episode
dependson knowledge of changing channel dimensions.The
Paper number 3W0283.
0043-1397/83/003W-0283$05.00
643
644
GRAF'VARIABILITY
OFSEDIMENT
REMOVAL
USA
ARIZONA
PhOenix
/nut
Gulch
Intensi•
Tombstone
I
Am
II
3
Fig. 1. Walnut Gulch Watershed, southernArizona.
dimensionsin turn provide the basisfor the analysisof sediment volumes removed during the erosion episode. Field
checksshowthat channelerosionduring the pastcenturyis not
likely to haveincisedbedrockbut ratherhas excavatedalluvial
deposits.Therefore,if the dimensionsof the small,pre-erosion
channelsand of the enlargedpost-erosionchannelsare known
at various cross sections, the differences between areas under
the two crosssectionsrepresentthe amountsof materialeroded
(Figure 2). Data for channel cross sectionsfrom the posterosionperiod were availablefrom 43 surveyedcross-channel
profilesmade in 1961 along a 13-km (8-mi) reachof'the main
stem of Walnut Gulch. The profile data are stored at the
SouthwestRangelandWatershedResearchCenter in Tucson,
Arizona. Data for the channel cross sectionsfrom the preerosion period were from the General Land Office Survey
recordsmade in 1881 and 1905,with the majority for the later
date. Plat mapsand surveyor'snotesare availablein the Phoenix, Arizona, officeof the Bureau of Land Management for 55
crosssectionsscatteredthroughout the basin.Although more
data would be useful,the historical record is limited.
The cross-sectionaldata for sediment removed by erosion
were extended into the third dimension along channels
throughoutthe basin.For an analysisof the channelsof the
entire basin,the data were organizedaccordingto the Strahler
stream order method [Doornkamp and King, 1971]. The
channelordersusedwere from a previousstudy by Murphy et
al. [ 1977],permittingthe calculationof the amountof sediment
removedfrom an averagecrosssectionfor eachorder.The total
amount of sedimenteroded from the system(which is seventh
order)wasrepresented
by the function.
S,- Y',/_...q.
(1)
where
S, total volumeof sedimentremovedfrom the basinchannel
system;
u stream order;
n maximum order;
L,
S,
total lengthof streamsof order u;
mean crosssectionalarea of sedimentremovedby erosion
on streams of order u.
Calculation
of the total volume of sediment removed from
the highestorder streamwas possiblein greaterdetail than an
overallaveragebecauseof numerouscrosssectionson themain
stream. The total reach was divided into 43 subunits,with each
subunit centeredon a measuredcrosssection(seeFigure 1 for
the location of the reach of intensivestudy). The amount of
sediment removed from the total reach then was
"
= Y,
(2)
$=1
where
1961
Surface
• Bedrock
'f½,'/½•
Alluvium
S,
order channel;
s
.... 1905
Surface
• Alluvium
Removed,
1905-1961 m
Fig. 2. Schematicdiagram showinga typical crosssectionon the
main streamof Walnut Gulch and illustratingthe methodof determining sedimentvolumeerodedfrom the crosssection.Not to scale.
total sedimentvolume eroded from the seventh(highest)
number ofcross section'
maximum number ofcross section(43 in this case);
S• crosssectionalarea of sedimentremovedfrom crosssection s'
L• lengthof the subunitcenteredon crosssections.
GRAF: VARIABILITY OF SEDIMENT REMOVAL
Ds-1
Do
Ds
Ds+l
I
I
S-1
I
where•:cis criticaltractiveforcein Newtonsper meterssquared
and d is the medianparticlediameterin millimeters.The importance of critical tractive force to the present study is that
tractive force generatedby someflows may be insufficientto
move commonly found particle sizesin some reachesof the
I
S
S+1
I
Da
645
channel.
Db
RESERVATIONS
Ls
Fig. 3. Schematic diagram defining factors used to calculate
volume of sedimenttransportedduring the erosionepisodeon Walnut
Gulch.
Each of the surveyedcrosssectionshad two typesof data' Ss,
cross-sectional
area of lost sedimentand Ds, downstreamdistance of the cross section from an arbitrary starting point
(Figure3).The boundariesof eachsectionwerelocatedhalfway
betweenthe measuredcrosssections,so that the length of the
end subunitsrequiredfor the solutionof (2) was
L• = D• +
L m=
D2--D •
2
Dm-- Dm- 1
2
(3)
(4)
Two measuressupply potential processinformation in the
followinganalyses.First is the conceptof 'tractive force' suggested,named,and definedby DuBoys in 1879 [Bogardi, 1978,
p. 80] where
The resultsof the investigationof spatialand temporalvariation of sedimenttransport must be viewed with certain constraintsin mind. The channeldimensionsfor the period before
the erosionepisodeare from the notesof surveyorswho were
more concerned with township survey than with stream
channels.General Land Office surveysthat can be checkedin
other areas by independentevidencehave proven reliable in
some casesbut not in others [Cooke and Reeves, 1976]. The
recordsin the Walnut Gulch area appear reasonablein light of
photographicevidenceand recollectionsof long-time residents
[Hastingsand Turner, 1965].
Also, the original surveysdo not provide samplesof channel
dimensionsas frequentlyalong the main stream a• do later
surveys.The lack of depth information at many crosssections
in the older surveysrequired the use of depth dimensions
calculated from allometric relationships established from
known crosssections[Bull, 1975; Graf, 1979], further reducing
the precisionof the data. Channeldimensiondata from the old
surveysrepresenta definiteimprovementover qualitativeestimates, however.
The estimateof sedimentremovedduring the erosionepisode
involved comparisonof the pre-erosionchannelswith those
surveyed in 1961. Most of the erosion probably occurred
% = 7RS
(5) during eventsof intenseprecipitationin the early 1930'sand
early 1940's(precipitationdata publishedby Cookeand Reeves
where
[1976, pp. 65-79]). An implicit assumptionwasmade that after
the episodeof erosionwascompleted(by 1945 at the latest),no
z0 meantractiveforce,N m-2,
channelenlargementoccurred.This assumptionwas probably
7 unitweightof water(9807N m- 3);
not strictly met, and some bank or bed scour probably ocR mean hydraulicradius,m;
curred after the erosionepisodeand before 1961. The amount
S energyslope(mustbe assumedto be equal to bed slopein
of material involved, however,is not likely to have altered the
field applications).
calculationsappreciablybecausethere were few significantpreIt is perhapsmisleadingto referto the DuBoys valueas tractive cipitation episodesduring the period [see Cooke and Reeves,
force sinceit is not a force in the physical-mathematicalsense, 1976,pp. 70-72].
but the term and equationare widely accepted[Bogardi, 1978,
In the comparisonof rates of process,averageyearly values
p. 83]. The geomorphicsignificanceof tractive forceis that it is were usedfor convenience,but actually the processes
are disdirectly related to the competenceof streamflow over a wide continuous,and year to year variation is substantial.Along the
range of particle sizes (recent reviews by Baker and Ritter channel,periodsof scourand fill alternatewith eachother. The
[1965] and Church[1978]). The units for tractive force(e.g.,as average rates of sedimentyield from slopesdo not take into
shownby Church[1978, p. 756]) are the sameas the units for account climatic variability but submergesuch variability in
shearstress(asgivenby Stelczer[1981, p. 18]).
long-term averages.Sediment production rates are likely to
The second measure of potential processis total stream have been significantly different in the period prior to the
power, the amount of power expendedby flowing water per erosion episode when the vegetation cover was different in
unit lengthof channel.As definedby Bagnold[1966, 1977] it is
many parts of the basin.Variation in human-inducedchanges,
f• = 7QS = %VW = 7RSVW
(6) suchas road building, mining, and grazing of cattle, also limit
the extension of estimates of production of sediment from
where f• is total streampower in N, V is mean velocity of flow slopes.
in m s-1, W is widthof flow in meters,and othersymbolsas
before.The geomorphicsignificanceof total stream power is
that it is directly related to total sedimentdischarge,that is,
stream capacity assuminga constantsupply of sediment(see
GRAF [1971] for a review).
Church[1978] found that the critical tractive force for very
loose sediment was
•:c= 1.78 d
(7)
SPATIAL VARIATION
The total
amount
of sediment
removed
from the channels
was enoughmaterial to coverthe basinto a depth of 4 cm (1.6
in). The valuesreportedin Table 1 showthat the major portion
of sediment
removed
from
the channels
of Walnut
Gulch
Watershed was removed from the highest-orderchannel. Despite the great numbersof lower-orderchannels,the relatively
646
GRAF' VARIABILITY
OFSEDIMENT
REMOVAL
TABLE 1. SedimentVolumesRemovedDuring ErosionEpisode,Walnut Gulch Watershed,SouthernArizona
Mean
Stream
Stream
Mean
Mean
Stream
Length,
Drainage
Area,
Mean
Cross-Sectional
Area,
Individual
Volume,
Total
Volume,
Volume,
% of
Order
Number
m
km2
m2
m3
m3
Total
1
2
3
4
5
6
7
Total
6134
1373
257
58
15
3
1
96
190
508
1,306
2,430
5,712
13,051
0.01
0.03
0.13
0.50
1.95
7.62
146.52
0.1
0.2
0.8
2.4
7.3
22.2
*
10
38
406
3,134
17,739
126,806
*
61,300
52,200
104,000
182,000
266,000
380,000
5,100,000
6,150,000
1
1
2
3
4
6
83
100
Table values in the final column are rounded.
*Mean valuesreplacedby detailed surveydata.
small amount of material removed from each one results in low
amountsof sedimentbeingerodedfrom them in total. Whether
or not theseresultsare typical of semiaridwatershedsremains
to be seen,becausedata are not availablefor comparison.In a
similar analysisof basinsin a humid region,J. C. Knox (Universityof Wisconsin,Madison,personalcommunication)found
that most of the sediment lost from channels in an erosion
episodecamefrom the lower-orderstreams.
The spatialvariationof erosionis directlyrelatedin budgetary fashion to a comparisonbetweentotal streampower and
the amount of sedimentsuppliedto the channelsystem.It is
impossibleto construct a complete time seriesof sediment
supplyfrom surroundingslopes,but the channelsare excavated
into•unconsolidated
materials,eitheralluviumdepositedby the
streamin its geologicallyrecentconfigurationor basinfill shed
from Surroundingmountains.Therefore,throughoutthe period
of interestsedimentsupply has exceededtransport capacity,
and spatialvariationin form is attributableto spatialvariation
in transportcapacityor total streampower.
The Walnut Gulch caseprobably differsfrom the southwestern Wisconsinexamplebecauseof the particlesizesinvolved.In
Wisconsin,sedimentsare from fine-grainedloesssoils(Figure
4) so that they are carried in suspensionand bedload is not a
largeproportion of the total load (Schumm[ 1977] reviewsriver
transport types and regions).In Walnut Gulch much of the
sedimentis coarsecontinentalalluvium carried by the streams
asbedload.The particlesized8,•for the alluviumis in the 4.0- to
8.0-mm range [Osterkampet al., 1982]. Field investigationrevealedmany particles25 mm or larger in mediumdiameter.In
lower ordersdepth of flow is insufficientto generatethe high
valuesof tractiveforce(or competence)
requiredfor transport.
The large amounts of sedimenteroded from the highestorder of the channelnetwork(as shownin Table 1 and Figure
5) direct attention to a detailedanalysisof the main streamof
Walnut Gulch. A plot of crosssectional area of sedimentremoved by erosion againstdownstreamdistancein the trunk
stream (Figure 6) showsthat the amount of material removed
first increasesand then decreases
downstream.Generalexplanation of the trendsin Figure 6 dependson sedimentsupply,
tractive force,and total streampower.
As statedpreviously,sedimentsupplyhasexceededtransport
500.0 -
100.0
50.0
100-
E
ß-• 80
(o
10.0 -
x
Walnut
._o 60,
.....
i-
•
LoessWatershed,
5.0
"'
ß
-
Gulch
•
/
o
40-
•
o
.g
1.0
/'
0.520-
0
0.01
0'.05
0:5
5'.0 0'.0
Grain Size, mm
Fig. 4. Particle size analysesfor the channelsof the main trunk of
Walnut Gulch and a typical loesswatershedin the midwesternUnited
States.
0.1
, /i
Order
Fig. 5. Distribution by streamorder of mean area of channelcross
sectionexcavatedby erosionin channelsof Walnut 6ulch Watershed.
GRAF' VARIABILITY OF SEDIMENTREMOVAL
Distance
0
m
5,000
I
1200
1000
Downstream,
647
10,000
I
15,000
.I
-
-10,000 -1•
o
(3,)
-
E
o
(3,) 800 -
E
(3,)
-
E
.t
'O
600
03
E
-
-
-5,000
<• 400
<
._o
o
•
x
200
i
x
0 '
0
5,000
10,000
i
I
20,000
30,000
Distance
Downstream,
I
I
I
40,000
000
ft
Fig. 6. Downstreamdistributionof area of channelcrosssectionexcavatedby erosionon the main streamof Walnut
Gulch. SeeFigure 1 for the locationof thisintensivelystudiedreach.
capacity throughout the past century and is not a variable in
this discussion.Generally, as higher orders are considered,the
10-year dischargeat any one crosssectionincreases,channel
slope exhibits relatively small declines, and width increases
dramatically(Figure 7). The interactionof dischargeand width,
however,causesa decreasein depth from sixth to seventhorder
(Figures 6 and 7). The amount of material excavatedfrom the
upper reachesof the network is relatively small where depth of
removal include a peak of sediment removal near 4000 m
(about 14,000ft). Sedimentarymaterialsin the reach are dominated by fine-grained,easily entrained materials, so that the
arroyo left after the erosionepisodewassignificantlywider and
deeperin this reach than in nearby reaches.The wide fluctu-
flow is insufficient
stances, areas where little erosion occurred had stored little
to entrain
the basin materials.
In midbasin
ations
in amount
of material
removed
from
the lowermost
reachesresultfrom the influencesof geologicstructureand the
nature of the sediment transport processitself. In some in-
areas,tractive force(competence)is sufficientto entrain materials and the channel is wide enough to permit high values of
total power (capacity),so large amounts of material were lost.
In the lowestreaches,the channelbecameso wide that depth of
sedimentinitially becausethe channel was eroding through
exposedresistantstrata. In such placesthe channel was relatively narrow, deep, and had a steep gradient. Other fluctu-
flow and tractive
reachesreflect massesor pulsesof sedimentsmoving through
the system.Becausethe sedimentsare moved by discontinuous
events,thesepulsesare probably temporary features.
force declined and limited
the amount
of
ations in the amount
of sediment removed from the lowermost
material lost by erosion.Transmissionlosses,unaccountedfor
in this analysis,also depletedavailable dischargeand tractive
force,as Wellas total streampower [Renard,1970;Lane et al.,
TEMPORAL VARIATION
1980].
The final distribution of erosionis modified by a variety of
The amount of material eroded from channelstorageareas
local influencesextendinga few hundredsof meters along the during the erosionepisodecan be placedin a temporal context
channels in a variety of locations. These local phenomena by comparisonwith data collectedin other studiesof Walnut
representvariability in bank resistanceso that the generaltrend Gulch. Comparisonsare possiblefor channel storage on a
of changing amounts of channel erosion in the downstream several-thousand-year
scaleand on a scaleof a few yearsduring
direction, as shown in Figure 6, is imperfect.At downstream the post-erosionperiodfor sedimentfrom slopesand channels.
distance8230 m (27,000ft), a minimum of erosionindicatesthat
Although large volumes of materials are involved in the
relativelysmall amountsof material were removedin compari- recenterosionepisode(about 6.2 x 106 m3), the volumeis
sonto adjacentchannelareas.The area of the reducedsediment small in comparison to the amount of alluvium remaining
lossis spatially coincidentwith surfaceoutcropsof Schieffelin beneath the channels. A well in the lower reach of the main
Granodiorite (Tertiary) and Naco Limestone(Permian).These stream channel revealedthat alluvium extendedto a depth of
highly resistantmaterialscausemore narrow, deep,and steep 29 m (95 ft) below the surfaceof the modern channel[Renard,
channelmorphologythan the Tertiary-Quaternaryalluvial ma1977]. Therefore, based on survey data of channel crosssecterials that dominate the other reachesof the main stream (for a
tions at a similar site nearby and assumingcontinuity of subgeologicmap of the basin, see U.S. Departmentof Agriculture surfaceconditions,the episodeof erosionremoved only about
[1970]). Channelsthrough the resistantsectioncontainedless 15% of the alluvial materials stored in the lower reaches. Even
alluvium for potential erosionthan other nearby reaches.
allowing for substantialentrenchmentof the channeland lesser
Other specificdetailsof the spatialdistributionof sediment depthsof alluvium upstream,the amount of material removed
648
GRAF' VARIABILITY OF SEDIMENT REMOVAL
100.0-
50.0-
b
./
10.0-
2.0-
5.0
1.0-
E
0.5-
1.0
.•'
0.5
Order
0.1
Order
d
200.0
100.0
E 5o.o
0.020-
(•
ß
0.010-
ø
._•
2
,,-
0.005-
10.0
./.
0.001
Order
1.0
Order
Fig. 7. The distribution by stream order of mean values of hydraulic parametersin channels of Walnut Gulch
Watershed.
A, width' B, depth'C, channelbedslope'D, magnitudeof calculated10-yearpeakdischarge.
in the erosionepisodeaccountedfor only a minor portion of
sedimentremoval during the erosionepisodethereforewas 18
timesthe rate observedafter the episode.
Local accounts indicate that most of the catastrophic
Sedimentcontributionsto the channelsfrom surrounding
channelerosionoccurredwithin a 15-yearperiodcommencing slopesunder conditionsof the post-erosionperiod are about
about 1930.The erosionprocesswasinconsistentfrom one year equal to the amount of material transported through the
to the next, but the overall averagerate of sedimenttransport channels. In recent work, Sireantonet al. [1977, 1980] and
the total amount still stored in the basin.
from channelstoragewasabout434,000m3 per year.Studies Renard [1980] have calculatedthat 20,000 to 25,000 m3 of
by Renard [1972] and Renard and Laursen [1975] show that
during the post-erosionperiod the yearly averageof sediment
material are erodedfrom the basin slopes.Given the accuracy
of the estimatesand variablelengthsof record,the annualslope
outputfrom the entirebasinwasabout24,100m3. The rate of
erosion
can be considered
identical
to the annual
channel
GRAF' VARIABILITYOF SEDIMENT
REMOVAL
Distance Downstream,
m
5,000
10,000
I
lOOO
649
15,000
I
I.
900 -
800 700
.,-,
600
"'"
500
200
• 400
100
3oo
2oo
lOO
0
i
0
10,•)00
i
I
20,000
i
'
30,•)00
40,•)00
'
-0
50,000
Distance Downstream, ft
Downstream
Distance, m
15,000
10,000
5,000
• -1.5
I
i
E
b
o
101000 '
20,•00
'
30,•)00 '
40,•)00
50,000
Downstream Distance, ft
Distance Downstream,
o
m
5,000
lO,OOO
I
15,000
I
I
150
E
lOO
c•
50
I
0
I
10,000
I
I
I
20,000
I
30,000
i
I
40,000
o
50,000
Distance Downstream, ft
Fig. 8. Downstream
distributionof hydraulicvariablesin the main trunk of WalnutGulchWatershed.
A, width' B,
depth;C, magnitudeof the 10-yearpeakdischarge.
Channelbedsloperelativelyinvariateandnot plotted.
transportof 24,100m3. If the presentratesof sedimentproductionfrom slopesand throughoutwereto be maintained,the
material excavatedfrom the basin during the erosionepisode
would not be replaced.
IMPLICATIONS FOR PRESENT PROCESSES
The spatial distributionof sedimentremoval left behind
channelformsthat haveimportantimplicationsfor continuing
processes.
Becausethe vastmajority of sedimentsin the semi-
arid streamsof Walnut Gulch are transportedas bedload and
becausein the main channellargeamountsof sedimentsremain
available for entrainment,tractive forceand total streampower
representreasonable
indicatorsof the ability of the channelsto
transport sediment[Graf, 1971]. When these measuresare
calculatedfor the dischargeof the 10-yearflood along Walnut
Gulch (flood data from previouswork, publishedin part by
Kniselet al. [1979] and ReichandRenard[1981]), the resulting
distributionsshow that presentconditionsare different from
650
GRAF: VARIABILITY OF SEDIMENT REMOVAL
20,000-
10,000-
indication that the coarse particles are not entrained in the
lower reaches.The large particlesrequirealmost45 N m2
Total
Stream
Power/
Force
....
tractiveforcefor motion (from (7) and 25-mm particles),and as
shown in. Figure 9, this critical level is not attained in some
Tractive
5,000
reaches.
In the main trunk of Walnut Gulch, downstream trends in
z
tractive force are much differentafter the erosionperiod than
they were before(Figure 10). An erratic but generaldeclinein
tractive force in the downstreamdirection replaceda marked
downstreamincrease.It appearslikely that the largestparticles
in the stream sedimentscan be carried in the upper reachesbut
not the lower reachesof the trunk stream,while the reversewas
1,000
5OO
500-
E
z
lOO
•
100-
o
5o
50-
.>
(,..3
10
10-
Order
Fig. 9. Distributions by stream order of mean valuesfor tractive
forceand total streampowerfor channelsof Walnut Gulch Watershed.
those prior to the erosion episode(see Graf [1982] for a detailed analysis from a different area). Figures 7, 8, and 9 show
the spatialdistributionof hydraulicmeasures.
Under conditions of the post-erosionperiod, total stream
power increasesconsistentlywith increasingorder (Figure 9).
Tractive force also increaseswith increasingorder until the
highestor seventhorder is encountered(Figure 9). Dramatic
increasesin width and concomitant declines in depth serve to
limit tractive force and competence(Figure 7). The largest
particlesin the channelmay thereforebe carried in midbasin
areas, but in lower reachesthey may not be transported because of lower
levels of tractive
force. Renard
and Laursen
[1975] noted the fine characteristics
of sedimenttransportedin
the lower reachesas opposed to upstream areas, a further
Downstream
true before arroyo development.
In the trunk, total streampower increasedin the downstream
direction more rapidly before the erosion episode than after
(Figure 11). The post-erosionsystemlacks the steepgradients
in total power found in the pre-erosion system,resulting in
more throughput of sedimentsin midbasinarea at present.In
the more recent channel system,well-definedboxlike channels
(which replacedwide shallow swales)produce higher levelsof
total power in the upper reachesof the main stream because
before arroyo developmentthe channelcould not contain all
the water in the 10-yearflow.
One of the most important implicationsof the spatial variation of sedimentremovalis that it imposesa particular spatial
control on subsequentfluvial processes.
The channelmorphology left after the erosion episode dictates the likely loci of
erosion and deposition. Since sediment is readily available,
those reacheswith sharply declining tractive force and total
power are likely depositionsites,while thosewith sharp increasesare likely erosion sites. The placement of one or a
limited number of recording instrumentsis therefore likely to
provide an inaccurateor misleadingrepresentationof the true
nature of processes
along the main stream.The temporal variation of sediment
also makes limited
recorded
Distance, m
5,000
10,000
I
120
removal
15 000
I
1905
100
1961
8O
6O
40
20
0
'
10,•)00
'
20,•)00
'
30,(•00
'
40,•)00
50,000
Downstream Distance, ft
Fig. 10. Downstreamdistributionsof tractiveforcefor themain trunk of Walnut Gulch.SeeFigure1 for locationof this
intensivelystudiedreach.For referencepurposesthe largestcommonparticleshavecriticaltractiveforcelevelsof about45
Nm-2
data
difficult to interpret, sincesuchdata are usefulin a predictive
senseonly until the next catastrophicadjustment.
The calculationsreported here rely on flow generatedby the
10-yearflood, which has a magnitudetoo small to substantially
affectthe channelmorphologyof the network within the span
of a singleevent. The continuedmovementof materials from
low-order streamsand storagein high-order streamsmight in
time alter the spatial arrangementsof tractive force and total
stream power, but the role of high-magnitudeeventsin sucha
GRAF: VARIABILITY OF SEDIMENT REMOVAL
Distance
o
25,000
•
Z
Downstream,
651
m
5000
10,000
15,000
I
I
I
20,000
-• O•15,000
'• Q"10,000
'•E
/
00
x•/
/
!
.... 1905
•
1961
I
I
I
I
10,000
20,000
30,000
40,000
Distance
Downstream,
50 000
ft
Fig. 11. Downstreamdistributionsof total streampower for the main trunk of Walnut Gulch. SeeFigure 1 for location of
thisintensivelystudiedreach.
schemeis unknown. The 100-year event, for example, might
completelychangethe presentsystemand establishnew spatial
arrangementsover a shortperiod.
CONCLUSIONS
Tentative answers are possible for the basic research
questions.First, rates of sedimenttransport during the erosion
episodewere about 18 times greater than channel sediment
transport rates or production of sedimentfrom slopesduring
the post-erosionperiod. The precipitationrecord [Cooke and
Reeves,1976,pp. 70-73] showsat leasttwo remarkableyearsof
intense precipitation during the episode.Second,during the
erosion episode beginning about 1930 the majority of sediments evacuated from the basin were eroded from the highestorder stream.At crosssectionsalong the lengthof the highestorder segment,the amount of sedimentremovedincreasedin
the downstream direction to a maximum and then declined, a
distributionreflectingthe conflictinginfluencesof variation in
dischargeand channelwidth. Third, the arroyo forms left by
the erosion episodecontrol the distribution of tractive force
and total streampower so that renewedstorageof sedimentis
likely in only a few limited areas.
A geomorphologicalperspectiveon the sedimenttransportation processeson the Walnut Gulch watershedas outlined
above indicatesthat short-termanalysesof the processes
produceaccuratebut impreciseviewson the processes
that are not
comprehensive.
Processratesobtainedfrom instrumentedportions of the basin during the past three decadescannot be
extendedspatially or temporally without risk of substantial
error. Without the short-term analyses,the longer-term geomorphologic perspectiveprovides a highly generalizedview
lackingin detail, especiallywith respectto presentlyoperating
processes.Evidence from the Walnut Gulch Watershed suggeststhat the prudent watershedanalystis one who employsa
judiciousmixture of engineeringand geomorphicapproaches
to basicand appliedresearch.
Acknowledgments.I am indebtedto Victor R. Baker (Department
of Geosciences,
University of Arizona) for indispensibleassistancein
unravelingthe tangledliteraturepertainingto streampower. Much of
the background developmentfor the researchreported in this paper
was supported by grant EAR-7727698 from the National Science
Foundation,Earth SciencesDivision,GeologyProgram.The research
on Walnut Gulch was financiallysupportedby a grant from the U.S.
Department of Agriculturefor a cooperativeagreementbetweenthe
SouthwesternRangeland Watershed ResearchCenter (Agricultural
ResearchService,Tucson)and the Department of Geography,Arizona
State University. The researchcould not have been accomplished
without the assistance of several individuals at the research center. Ken
Renard, ResearchDirector, and Leonard Lane, hydrologist,cooperated during the project and offered helpful suggestionsin the preparation of this report. Del Wallace, geologist,generouslyprovided
valuabledata for the study.
REFERENCES
Bagnold,R. A., An approachto the sedimenttransport problem from
generalphysics,U.S. Geol.Surv.Prof. Pap. 422-J, 21 pp., 1966.
Bagnold,R. A., Bed load transport by natural rivers, Water Resour.
Res., 13, 303-312, 1977.
Baker, V. R., and D. F. Ritter, Competenceof riversto transportcoarse
bedload material, Geol. Soc. Am. Bull., 86, 975-978, 1975.
Bull, W. B., Allometric changeof landforms,Geol. Soc. Am. Bull., 86,
1489-1498, 1975.
Bogardi,J., SedimentTransportin Alluvial Streams,Akademiai Kiado,
Budapest,1978.
Church, M., Paleohydrologicalreconstructions
from a Holocenevalley
fill, in Fluvial Sedimentology,
edited by A.D. Miall, pp. 743-772,
Canadian Societyof PetroleumGeologists,Calgary, 1978.
Cooke, R. U., and R. W. Reeves,Arroyosand EnvironmentalChangein
the American South-West, Clarendon, Oxford, 1976.
Doornkamp, J. C., and C. A.M.
King, Numerical Analysisin Ge-
omorphology,St. Martin's Press,New York, 1971.
Emmett, W. W., Channel aggradation in western United States as
indicatedby observationat Vigil Network sites.Z. Geomorphol.21,
52-62, 1974.
Ferreira, V. A., Bibliography: SouthwestRangeland Watershed Research Center, report, 51 pp., U.S. Dep. of Agric., Tucson, Ariz.,
1979.
Gilluly, J., General geology of central CochiseCounty, Arizona, U.S.
Geol.Surv.Prof. Pap. 281, 169 pp., 1956.
Graf, W. H., Hydraulics of SedimentTransport, McGraw-Hill, New
York, 1971.
Graf, W. L., Developmentof montane arroyosand gullies,Earth Surf
Proc., 4, 1-14, 1979.
Graf, W. L., Downstreamchangesin streampower,Henry Mountains,
Utah, Ann. Assoc.Am. Geographers,
in press,1983.
Hastings,J. R., and R. M. Turner, The ChangingMile, University of
Arizona Press,Tucson, Arizona, 1965.
Hunt, C. B., Natural Regionsof the United Statesand Canada, W. H.
Freeman, San Francisco, 1974.
Knisel, W. G., K. G. Renard, and L. J. Lane, Hydrologic data collection: How long is enough?,in Proceedings
from the Specialty
Conferenceon Irrigation and Drainage in the Nineteen-eighties,pp.
238-254, American Societyof Civil Engineers,Albuquerque,New
Mexico, 1979.
Lane, L. J., V. A. Ferreira, and E. D. Shirley,Estimatingtransmission
loss in ephemeralstream channels,H ydrol. Water Resour. Ariz.
Southwest,10, 193-202, 1980.
Leopold, L. B., Reversalof erosioncycle and climatic change,Quat.
Res., 6, 557-562, 1976.
Leopold, L. B., and W. W. Emmett, Bedloadmeasurements,
East Fork
River, Wyoming, Proc. Nat. Acad.Sci.,73, 1000-1004, 1976.
652
GRAF: VARIABILITY OF SEDIMENT REMOVAL
Meyers, J., The Last Chance: Tombstone'sEarly Years, Dutton, New
York, 1950.
Murphy, J. B., D. E. Wallace, and L. J. Lane, Geomorphicparameters
predict hydrographcharacteristicsin the Southwest,Water Resour.
Bull., 13, 25-38, 1977.
Osterkamp, W. R., E. R. Hedman, and A. G. Wiseman, Geometry,
basin characteristics,discharge,and particle-sizedata from gaged
stream-channelsites,Western United States, U.S. Geol. Surv. Open
File Rep.82-93, 56 pp., 1982.
Reich, B. M., and K. G. Renard, Application of advancesin flood
frequencyanalysis,Water Resour.Bull., 17, 67-74, 1981.
Renard,K. G., The hydrologyof semiaridrangelandwatersheds,
A•lric.
Res.Serv.Rep.ARS 41-162, 26 pp., 1970.
Renard, K. G., Dynamic structureof ephemeralstreams,Ph.D. dissertation, Dep. of Civ. Eng. and Eng. Mech., Univ. of Ariz., Tucson,
1972.
Sellers,W. D., and R. H. Hill, Arizona Climate,1931-1972,University
of Arizona Press,Tucson, 1974.
Schumm,S. A., The Fluvial System,Wiley-Interscience,New York,
1977.
Shen,H. W., Modelingof Rivers,JohnWiley, New York, 1979.
Simanton, J. R., H. B. Osborn, and K. G. Renard, Effects of brush to
grassconversionon the hydrology and erosionof a semi-aridsouthwestern rangeland watershed,Hydrol. Water Resour.Ariz. Southwest, 7, 249-256, 1977.
Simanton,J. R., H. B. Osborn, and K. G. Renard, Application of the
USLE to southwesternrangelands, Hydrol. Water Resour. Ariz.
Southwest,10, 213-220, 1980.
Stelczer, K., Bed-Load Transport: Theory and Practice, Water ResourcesPublications,Littleton, Colorado, 1981.
U.S. Department of Agriculture, Soil survey, Walnut Gulch experi-
mental watershed,Arizona, spec.rep., 55 pp., U.S. Dep. of Agric.,
Tucson, Ariz., 1970.
Renard, K. G., Past, present,and future water resourcesresearchin
arid and semiarid areas of the southwesternUnited States, paper
presentedat the Hydrology Symposium,Inst. of Eng., Brisbane,
Wilson, E. D., A resumeof the geologyof Arizona, Ariz. Bur. Mines
Bull. 171, 140 pp., 1962.
Australia, June 28-30, 1977.
Renard, K. G., Estimating erosionand sedimentyield from rangeland,
paper presentedat the Symposiumon WatershedManagement,Am.
Soc.of Civ. Eng., Boise,Idaho, 1980.
Renard, K. G., and E. M. Laursen, Dynamic behavior of ephemeral
streams,,I. Hydraul. Div. Am. Soc.Civ. Eng.,101,511-528, 1975.
(ReceivedAugust4, 1982;
revisedFebruary 1, 1983;
acceptedFebruary 16, 1983.)