The University of Notre Dame
Invertebrate Resistance and Resilience to Intermittency in a Desert Stream
Author(s): Emily H. Stanley, Derek L. Buschman, Andrew J. Boulton, Nancy B. Grimm, S. G.
Fisher
Source: American Midland Naturalist, Vol. 131, No. 2 (Apr., 1994), pp. 288-300
Published by: The University of Notre Dame
Stable URL: http://www.jstor.org/stable/2426255 .
Accessed: 19/07/2011 18:51
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=notredame. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
The University of Notre Dame is collaborating with JSTOR to digitize, preserve and extend access to American
Midland Naturalist.
http://www.jstor.org
Am. Midl. Nat. 131:288-300
InvertebrateResistanceand Resilience to
in a Desert Stream
Intermittency
EMILY H. STANLEY', DEREK L. BUSCHMAN2, ANDREW J. BOULTON3,
NANCY B. GRIMM ANDS. G. FISHER
Tempe85287
DepartmentofZoology,ArizonaState University,
ABSTRACT. Invertebrateresponsesto water loss were investigatedduring drying,dry
Sonoran Desert stream. Some
and rewettingphases in Sycamore Creek, an intermittent
taxa were resistantto dryingbecause they could toleraterapidly changingwater quality
and/or move upstreamto avoid stranding.Communityshiftsoccurredat one site when it
became isolated fromup- and downstreamreaches; taxa such as beetles,hemipterans,and
thesnail Physellavirgatabecame dominant.No communitychangeswere detectedat a second
sitewhichremainedconnectedwithupstreamreachesby surfaceflow.Mortalityafterwater
loss was severeas few individualssurvivedlongerthan 10 days. Low resistanceduringthe
dryphase was associated with rapid moistureloss fromsediments.Invertebratepersistence
reacheswas due to recolonizationafterrewetting;however,densityincreases
in intermittent
afterfloodswhichreestablishedflowat drysiteswerelowerthanreportedvalues forperennial
sitesin SycamoreCreek. Slower ratesof recoverymay reflectthe composition,reducedsize
persistencein
colonistpools. Nonetheless,invertebrate
and remotenessof macroinvertebrate
desertstreamswhere both floodingand dryingare frequentis due largelyto theirability
to rapidlyrecolonizedisturbedsites.
INTRODUCTION
In temporary streams, drying is a key environmental feature influencing distribution,
abundance and life histories of the aquatic fauna (Boulton and Suter, 1986; Stanley and
Fisher, 1992; Boulton and Lake, 1992b). Intermittent conditions have led to a variety of
physiological (e.g., Crowe, 1971; Tauber et al., 1984; Berra et al., 1989) and behavioral
(Bouvet, 1977; Boulton, 1989) adaptations. Survival during periods of declining and disappearing discharge is a result of either avoidance or tolerance of changing environmental
conditions. Physiological tolerance to habitat water loss is well documented in organisms of
temporary ponds and pools (reviewed by Wiggins et al., 1980; Williams, 1985), but also
occurs among inhabitants of temporary streams (reviewed by Williams, 1987). Avoidance
may be behavioral (e.g., burrowing into substrates, Williams et al., 1974), or may involve
life history modification during drying (Taylor, 1983; Delucchi and Peckarsky, 1989),
emigration (Momot, 1966), or recolonization after disturbance (e.g., Williams, 1977; Sagar,
1983; Benzie, 1984).
Discharge in desert streams is characterized by flash floods and drying. Both disturbances
invariably influence life histories and community dynamics of the aquatic biota. While much
attention has been given to invertebrate responses to flooding (Gray, 1981; Gray and Fisher,
1981; Fisher et al., 1982; Meffe and Minckley, 1987; Grimm and Fisher, 1989), far less is
known about the effects of drying on macroinvertebrates in these ecosystems. Given the
widespread occurrence of intermittencyin desert streams and the importance and conspicPresentaddresses:
I Departmentof Biological Sciences,Universityof Alabama, Tuscaloosa 35487
2
Departmentof Philosophy,Purdue University,West Lafayette,Indiana 47907
3Department of Ecosystem Management, Universityof New England, Armidale 2351, NSW
Australia
288
1994
STANLEY ET AL.: RESISTANCE
AND RESILIENCE
289
in these systems(Fisher and Gray, 1983), we wanted to
uousness of macroinvertebrates
coped withdrying.This was done by addressingthe following
determinehow invertebrates
questions:(1) what general communitychangesoccur duringdrying,and (2) what are the
desert
mechanismsresponsiblefor persistenceof invertebratepopulationsin intermittent
streams?Persistenceovertimeis a productof resistanceto drying(the percentageof change
which occurs during the period over which the systemis rapidly shrinkingbut still wet;
resistanceI), resistanceafterwaterloss (changeswhichoccurduringsurfacewaterabsence;
resistanceII) and recoveryafterrewetting(resilience) (Stanley and Fisher, 1992). Field
surveysand experimentswere used todetermineresistanceand resilienceofaquatic macroinSonoran Desert stream.
vertebratesin SycamoreCreek, an intermittent
STUDY
SITE
SycamoreCreek arises in the Mazatzal Mountains and flowssouthto the Verde River
in Arizona's SonoranDesert. Elevationin themid-basinarea wherethisstudywas conducted
is 610 m and precipitationaverages 300-400 mm annually, fallingmostlyduringwinter
storms(December throughMarch) and summermonsoons(late July-September)(Thomsen
muchofthebasin during
1968). Streamflowis oftencontinuousthroughout
and Schumfann,
and over 50% of the channel
winter;however,in summerthe creek becomesintermittent
can lose surfaceflowduringdryyears (Stanley,1993). Dryingmay begin as earlyas April
precipitationand reducedevapotranspiand continuethroughoutthe year until sufficient
rationreversethetrendofwaterloss. Withina singlekilometertheremay be bothperennial
sitesand sitesthatflowas brieflyas threemonthsannually.The mid-basinarea ofSycamore
Creek is characterizedby low gradient,unshaded gravel runs and pools interspersedwith
rockyriffles.Sedimentsare dominatedby coarse sand and pebbles (5-50 mm) which may
reachdepthsof 200 cm in someareas, but are typically50-100 cm deep (Valett et al., 1990).
Summerwater temperaturevaries between20 and 30 C.
Routine invertebratesampling was conductedat two gravel reaches during 1989 and
1990. Site I was a large (140 x 20 m) reach whichmaintainedflowbetweenFebruaryand
June in 1989 and betweenlate March and June in 1990. Drying was characterizedby
disappearance of flow at the downstreamend of the reach followedby loss of upstream
connection,creatingan isolated pool beforecompletewater loss. Site II (122 x 15 m)
typicallyflowedforat least 7 months,and oftendried only partially.Drying at this site
was characterizedby loss of flow fromthe downstreamend of the reach in mid-summer
(June-July),thena rapid upstreamrecessionof surfacewater.In 1990, flowintotheupper
thirdof the reach persistedthroughoutthe summer;howeverin 1989, dryingof this reach
was completeas the upstreamrecessionof water progressedintothe riffleabove thisstudy
site.
METHODS
Duringdrying.-Macroinvertebratecommunitychangesduringwater loss were assessed
by randomlycollectingfivecore samples (core size = 80 cm2 insertedto a depthof 10 cm)
fromwettedareas of both sitesI and II in 1989 and site I in 1990. Samples were collected
everytwo weeks beforethe startof drying,then weekly afterflow became intermittent.
Samples were elutriated,filteredthrougha 250 ,umnet, preservedin 70% ethanol in the
field,thenlater sortedand identifiedin the laboratoryunder x 10 magnification.
activelyavoided ongoingdrying,trapswere constructedto
To determineif invertebrates
collectindividualsmovingup- and downstreamat site II. Traps consistedof funnels(20
cm taperingto 5 cm in diameter)to which 50 ,ummesh bags were attached.Three pairs of
traps were placed 1 m apart along a transectperpendicularto stream flow. Each pair
290
THE
AMERICAN MIDLAND
NATURALIST
131(2)
consistedof an upstream-facing
trap and a downstream-facing
trap. The lower halves of
the trap openingswere gentlypushed into the sedimentsjust above the sink (the point at
which surfacewater disappeared) and leftin place for28 h on 5-6 August 1990. Mesh
bags were replaced every4 h, and theircontentsrinsedout and preservedin 70% ethanol.
Sink locationwas also recordedat the timeof sample collectionas its positionwas first3
m below, thenlater 4 m above the traps. Differencesin abundance of individualtaxa and
total numbersof individualsmovingup- or downstreamwere determinedusing KruskalWallis tests(Zar, 1984) withBonferroni-corrected
levels(i.e.,levelofsignificance
significance
=
0.05/number of tests).
Afterwaterloss.-Stream sedimentswere collectedwithan 80 cm2core samplerfromfive
reaches (sites I and II plus threesimilarreaches located betweensites I and II which had
also lost surfaceflowduringMay and June). Water loss was monitoredcloselyin orderto
ensurethatsampleswerecollectedfroman area in whichtimeofdryingwas knownprecisely,
then2 coreswere collectedfromeach site 1, 3, 6, 10, and 15 days afterdryingto determine
invertebratesurvival afterdisappearance of surface water. A small subsample was also
taken fromeach core forsedimentmoisturedetermination.Samples were returnedto the
laboratoryin plasticbags and floodedwithdistilledwater.After24 h ofinundation,sediments
were elutriatedand filteredthrougha 50 ,umnet. Viable organismswere collectedand
identifiedunder a x 10 dissectingmicroscope.
Afterrewetting.
-To determineif macroinvertebrates
were able to rapidly recolonize
previouslydrystreamreachesafterrewetting,
samples were collectedfromsiteI withintwo
days afterrewettingon threeseparate dates in 1990. Samples were collectedon 30 March
followinga large winter flood which rewettedextensiveareas of previouslydry stream
channeland aftertwo summerfloods(17 Augustand 5 September)whichcaused temporary
rewettingof this site. Five sedimentcores were taken randomlywithinthe wettedstream,
preservedin 70% ethanoland returnedto the lab foridentification
and enumeration.Four
to six litersof water was also collectedwith a small bucketand filteredthrougha 250 ,um
net to gatherlarger,more mobile individualsfromthe water column.
Resilience (rate of recoveryfollowingdisturbance;Websteret al., 1974) was measured
on threeoccasions: followingthe March 1990 floodat site I and at site II, and duringa
periodofongoingdryingat site II in 1989. Resiliencewas measuredas the slope of the line
of invertebrate
densityovertimesince flooding.Density data were log(x + 1) transformed
to meet assumptionsof the linear regressionmodel, then differencesin resiliencewere
assessedbya multiplecomparisonofslopesofthetransformed
data versustime,and Tukey's
HSD testwas used to identifysignificantly
different
lines (Zar, 1984) at a significancelevel
of 0.05.
RESULTS
Invertebrate
resistance
todrying.
-There wereno consistentchangesin invertebrate
density
associatedwithdryingduring 1989 or 1990 at eithersite (Fig. 1). Effectsof dryingat site
II were confoundedby a floodon 23 July which significantly
reduceddensities(Student's
t-test,t = 4.60, P < 0.001). Postfloodrecoverywas negligibleas therewas no significant
difference
betweendensities2 days afterand 6 weeks afterflooding(Student'st-test,t =
0.65, P > 0.50), and unlike other postfloodsequences (Grimm and Fisher, 1989), days
sincefloodwas not significantly
correlatedwith invertebrate
numbers(Table 3).
There were distincttaxonomicshiftsduringthe week precedingcompletewater loss at
site I, but not at site II. Althoughthere were no changes in numberof taxa over time,
macroinvertebrate
communitiesbecame dominated by winged insects (coleopteransand
hemipterans)and the snail Physellavirgata(Table 1). On the finalday of dryingat site I
STANLEY ET AL.: RESISTANCE
1994
291
AND RESILIENCE
Site I
200 - 1 989
A
100t
E
\
0-
5/12
0n
50
0
140 -
030
20
5/22
6/5
6/9
9
a 10
5/5
Z
6/8
5/19
6/18
120* Site 1
40
5/27
6/26
Date
8/3
9/5
FIG. 1. Mean invertebratedensitiesduring dryingat site I in 1989 and 1990 and at site II in
1989. The firstdownward-pointingarrow in each panel indicatesthe approximatedate when flow
became discontinuous;the second downward arrow signifiesloss of upstreamflow into the site (site
I only). Upward-pointingarrow indicatesthe timingof a floodat site II. Error bars represent?1 SE
in 1989, snail densities were so high that an area of ca. 0.3 m2 was completely filled with
snails to a depth of 30 mm. Also abundant was the midge Chironomus, which is distinctive
because of its bright red color caused by presence of hemoglobin, enabling it to tolerate low
dissolved oxygen concentrations (Erikson et al., 1984). Mayflies, which had been common
before drying became extreme, were rare or absent this time.
A total of 2397 individuals from 27 taxa were captured moving up- or downstream during
drying at site 11 (1727 individuals from 23 taxa moving upstream, 670 individuals from 19
taxa moving downstream) (Fig. 2). No individual taxon showed significantupstream movement, but there was significant upstream movement for total number of individuals collected
292
THE
AMERICAN MIDLAND
131(2)
NATURALIST
TABLE l. Percentage of snails (Physella virgata) and winged adult Coleoptera and Hemiptera
presentin benthicsamples duringdryingat two sites
(P
=
Date
Streamsurface
area (m2)
% snails
% wingedadults
Site 1 1989
12 May
22 May
27 May
5 June
9 June
858
536
309
43
1
0.5
1.6
2.6
11.9
68.2
1.6
1.5
0.9
10.6
3.4
Site 1 1990
30 March
2 April
6 April
14 April
22 April
5 May
19 May
2 June
8 June
15 June
18 June
.308
263
043
724
708
706
686
495
317
64
2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
3.4
17.9
51.8
0.0
0.0
0.0
0.0
0.0
1.1
2.4
15.9
15.1
43.4
21.5
Site 11 1989
27 May
12 June
26 June
10 July
24 July
28 July
3 Aug
11 Aug
22 Aug
8 Sept
764
754
518
210
684
414
186
125
117
40
13.0
21.7
13.5
2.2
3.9
0.4
0.0
0.8
0.0
0.1
0.0
1.7
0.4
2.4
0.0
1.5
0.0
0.6
2.0
0.9
0.0003). During the monitoring period, sink location fluctuated from 3.5 m below the
traps at 0800 h to 4.7 m above the traps at 1600 h, exposing them between 1230 h and
2000 h; however,individualswere still capturedduringthis period. Greatestinvertebrate
movementin bothdirectionsoccurredbetween2400 h and 0400 h, while fewestindividuals
movedup- or downstreambetween 1600 h and 2000 h. In all but one sampling period
(2000-2400 h), netmovementwas upstream(Fig. 2). In contrasttotheseresults,downstream
movementat a perennial site in 1979 was significantly
greaterthan upstreammovement
(10,205 driftingdownstreamversus 594 movingupstream;P < 0.001, Wilcoxon paired
sample test) and forall 3 h samplingperiod downstreamdriftwas greaterthan upstream
movement(Fig. 2; Gray and Fisher, 1981).
Invertebrateresistance afterwater loss.-Only
12 taxa were collected during field surveys
ofdryreaches(Table 2). Sites whichdried slowly(water loss occurringover 3-4 days) had
moretaxa and individualsthan rapidlydryingsites (completewater loss within2 h), but
fewindividualspersistedat any site beyond10 days and only 5 individualswere foundtwo
1994
STANLEY
1000
500
ET AL.: RESISTANCE
-
-
293
AND RESILIENCE
Drying Reach
Upstream
cn
0L
Downstream
cn
-500
-
I
I
I
12 am
12 pm
-o
i
12 pm
Permanent Reach
2000-
C:
I
I
Upstream
-2000
-00Downstream
-4000
12 pm
B
12 am
12 pm
Time of Day
capturedin traps movingupstreamand downstream
FIG. 2.-Total numberof macroinvertebrates
in drying(top) and perennial (bottom) reaches. Bottom panel is modifiedfromGray and Fisher
(1981). Note that 0400-0800 samples fromthe dryingsite were lost
weeks postdrying (Fig. 3, Table 2). There was a significant relationship between number
of individuals present and sediment moisture for all samples (r2 = 0.46, P < 0.05). Physella
virgata and tipulid larvae and pupae were the most abundant taxa after water loss. The
tipulids appeared to be completing larval development, as samples collected from days 1
and 3 were dominated by larvae whereas pupae were most common on day 6.
Resilience after rewetting. Number of taxa collected from benthic sediments and the
water column soon (?2 days) after floods which rewet site I varied from two taxa after a
winter flood on 30 March to seven taxa following a summer flood on 17 August and six
aftera flood on 5 September. Early successional assemblages were dominated by chironomids
and oligochaetes in March, but hemipterans and coleopterans were the most common taxa
collected after the two summer floods. Nonflying individuals present immediately after
rewetting in August and September were predominantly dipteran larvae (Chironomidae
and Ceratopogonidae). Over 70 Probezzia (Ceratopogonidae), many of which were final
294
THE
AMERICAN MIDLAND
131(2)
NATURALIST
20E
1:5 t
E
\
0
0
'-10
5
E
z
0
__
3
6
9
12
15
Days After Water Loss
froma
sediments
foundin reinundated
of benthicmacroinvertebrates
FIG. 3.-Average densities
was prolonged
(circles)during
rapidly(squares)anda sitewheredrying
sitewheredrying
progressed
?1 SE
July,1990.Errorbarsrepresent
instarlarvae, were collectedfroma singlecore sample in Augustjust 2 days afterflooding
and flowresumption.
densitiesoccurredwithina monthafterrewettingat sites
Maximum benthicinvertebrate
I and II in March 1990 (Fig. 4). Recoveryratesfollowingthesetwo floodsand the autumn
floodat site II in 1989 were compared to assess differencesin resilience.Recoveryrates
among sequences (F = 12.84, P < 0.0005) as resiliencewas greatestat
varied significantly
site II afterrewettingin 1990, and lowest followingthe autumn floodat the same site in
1989 (Table 3).
DISCUSSION
ResistanceI.-Invertebrate densitiesmayincrease(Extence,1981), decrease(Boultonand
Lake, 1992b) or remainunchanged(Hynes, 1975; Chessmanand Robinson,1987) in streams
waterquality
duringdrying.Decreases have been attributedto intoleranceof deteriorating
(Boulton and Lake, 1992a, b) and intensifiedpredation(Larimore et al., 1959), whereas
habitats(Smith
in ever-diminishing
increasesare caused by mobileorganismscongregating
and Pearson, 1987). Like mostdryingstreams,physicochemicalconditionsshiftedrapidly
duringsampling,particularlyat site I (Stanley,1993), and undoubtedlyhad adverseeffects
on the benthos.For instance,sedimentsand surface water oftenbecame anoxic during
drying,favoringinvertebratesusing atmosphericoxygen (e.g., adult beetles,corixids) or
hemoglobin(Chironomus).Taxa with high dissolvedoxygenrequirements(e.g., mayflies)
STANLEY ET AL.: RESISTANCE
1994
AND RESILIENCE
295
TABLE 2.-Abundance of taxa collected in field surveysfromfivedifferentsites in dry reaches.
(+ = present[1-2 individualsper sample], C = common
Abundance is expressedsemiquantitatively
[3-101, and A = abundant [>10]). Larvae are denotedby L, pupae by P
Days since water loss
Taxa
1
3
6
10
15
C
C
C
+
+
Annelida
Oligochaeta
Crustacea
Copepoda
Ostracoda
C
+
+
Hydracarina
+
C
Coleoptera
Hydrophilidae(L)
+
+
Diptera
Tipulidae (L)
A
A
+
+
C
Hydracarina
+
+
Insecta
(P)
+
+
+
Chironomidae
+
+
Ceratopogonidae
Probezzia
C
A
+
+
+
+
+
+
+
+
C
C
+
+
+
+
11
9
Tabanidae
Tabanus
Gastropoda
Physidae
Physellavirgata
Planorbidae
Helisoma sp.
Total numberof taxa
10
+
7
5
were notably absent during the final days of drying at site I. Density decreases due to
predation are also likely given increased numbers of fish and predaceous beetles and bugs.
Benthic samples from site I were littered with insect body parts and empty snail shells
during the final days preceding water loss, and were likely remnants of predation. In many
temporary streams, densities of predators rise sharply during drying (Stehr and Branson,
1938; Hynes, 1975; Abell, 1984). Wiggins et al. (1980) suggested that predator recruitment
in temporary pools is timed to coincide with peak prey densities, and this may also apply
in intermittentstreams (Boulton and Lake, 1992b).
Observed upstream movement of invertebrates demonstrates that some individuals can
keep pace with drying. Similar behavior has been observed in other temporary streams
(Delucchi, 1989), but this ability was clearly limited. Although invertebrateschased receding
water margins upstream, drying was often faster than invertebrate movement, leaving
individuals stranded on the dry stream bed. Despite the absence of significant differences
THE
296
AMERICAN MIDLAND
131(2)
NATURALIST
120
E
100
o
zoq
8
4-
(7
o
-0
E
-
-
I/1
0 -
14
o
I
20-
z
.-.'-"
7
14
21
28
35
Days After Rewetting
densitiesfollowingrewettingof sites I (circles)
FIG. 4.-Postflood recoveryof macroinvertebrate
and II (squares), beginningon 30 March, 1990. Error bars are ?1 SE
in up- versus downstream movement within any taxon, some are clearly more vagile than
others. Failure to detect taxon-specific differences was likely due to small numbers of
individuals per taxon captured in traps. Community shifts at site I suggest that Physella
can follow receding stream margins, as can various beetle and bug species. Rather than
avoid drying pools, some beetles and hemipterans may in fact seek out such habitats as they
offerabundant food. For instance, Gray (1980) observed eight belostomatid adults (Abedus
herberti) fly into two adjacent drying pools and consume 20 fish within a matter of a few
hours.
In contrast to site I, there were no marked density or taxonomic changes at site II during
drying. This is likely due to the faster rate of drying and continued connection with the
upstream riffleat this site. Apparently, effectsof drying are detectable only when reaches
become discontinuous and isolated. Delucchi (1988) was also unable to find differences
densityvs. timeaftera rewetting
TABLE3.-Slopes oflines expressingchangesin macroinvertebrate
flood(sites I, II in 1990), a floodduringdrying(site II in 1989), and floodsat perennialsites (from
Grimm and Fisher, 1989). Values in parenthesesare standarderrors
Site
Slope
(SE)
Site 1 1990
Site 11 1990
Site 11 1989
0.76 (0.15)
2.82 (0.69)
0.39 (0.20)
Grimm and Fisher (1989)
Winter-spring
Summer
Autumn
1.33 (0.56)
1.37 (0.98)
2.45 (1.04)
r2
0.698
0.627
0.345
1994
STANLEY ET AL.: RESISTANCE
AND RESILIENCE
297
between temporaryand permanentstream communitiesat sites which never lost their
upstream connections.Communitychanges in isolated reaches, on the other hand, are
common,and theirdevelopmenthas been well documentedin a varietyof studies(reviewed
by Williams, 1987). Abell (1984) referredto this assemblage as the "summer clean-up
crew" as it is dominatedby scavengersand predators.
Resistance 1.-The lengthoftimeSycamoreCreek macroinvertebrates
are able to tolerate
waterloss is briefin comparisonto thosein temporarystreamsin Australia(Boulton,1989)
and southernCanada (Williams and Hynes, 1977). In SycamoreCreek, few macroinvertebrateshave restingstagesor use refugesduringdryperiods(Boultonetal., 1992b) resulting
in a virtual absence of resistanceto periods of water absence. This may reflectextreme
temperaturesof exposed sediments(>60 C; Stanley,1993), rapid eliminationof remaining
sedimentmoistureand sedimentscour upon rewetting(Gray, 1981). In most studies reportingdroughttolerance,invertebrates
are able to persistin moist refugessuch as algal
mats (Strandine,1941), crayfishburrows(Boulton,1989) or deep sediments(Clifford,1966;
Williams, 1977). Moist benthicrefugessimplydo not exist in drydesertstreamchannels.
Althoughhyporheicsedimentsare generallycooler and moisterthan surfacesediments
and oftenremain saturatedlong afterdisappearance of surfaceflow (Stanley and Valett,
1992), the hyporheiczone of SycamoreCreek is not an importantrefugeforbenthicspecies
(Boulton et al., 1992b). One possibleexceptionis the ceratopogonidlarva Probezzia,which
is commonin both benthicand hyporheiczones and was abundant immediatelyafterone
summer flood. It is unclear why the hyporheiczone does not serve as a refugefor the
benthos,particularlygiven the existenceof a diverse,obligate hyporheicfauna in some
Sonoran Desert streams (Boulton et al., 1992c). Althoughnumerous surface taxa were
collectedfromthe hyporheiczone in a South Australian arid-zone stream,Cooling and
Boulton (1993) feltthat these individualswere probablypassive colonistsand unlikelyto
be an importantsource of recolonistsafterrewetting.The high degree of specialization
neededto survivehotdryconditionsmaybe fargreaterthanforpersistencevia recolonization
(Hynes, 1975), and thephysicalconditions(e.g.,reducedoxygenand foodavailability,small
size of interstices)in the hyporheiczone may make it a poor refugeforbenthicmacroinvertebratesin drylandstreams(Gagneur and Chaoui-Boudghane,1991).
different
in the three
Resilience.-Recovery rates followingfloodingwere significantly
sequences examined in 1989 and 1990. Resilience was greatestaftera floodrewetsite II
in 1990, and least aftera summerfloodwhich occurredduringa period of dryingat the
same site in 1989. Recoveryratesat site I in 1990 and site II in 1989 were less than mean
recoveryvalues reportedforperennialreaches in SycamoreQreek (Table 3). Larimore et
al. (1959) suggestedthat recoveryin intermittent
streamsis a functionof the extentof the
area affected,
distancefroma colonistsource,or season,whereasBoultonand Lake (1992a)
foundthathistoricaleffects(e.g.,timingof waterloss the previousyear) stronglyinfluenced
recruitmentin two Australian temporarystreams.Rapid postfloodrecoveryin Sycamore
Creek is due to recolonizationby ovipositingadults withshortgenerationtimes(Gray and
Fisher, 1981).
We suggestthat slower recoveryrates in previouslydry sites are a resultof changes in
thecompositionoftherecolonizationpool. Dryingmayaffectthepool ofpotentialrecolonists
in threeways. First, assemblage structuremay shiftaway fromtaxa with aerial adults
(mayflies,chironomids,caddisflies)to one dominatedby snails (see Table 1 and Boultonet
al., 1992a). Secondly,habitat shrinkagereducesthe total numberof invertebrates
present
before flooding.Finally, dryingcan increase the distance between colonizationsources
sites. Althoughovipositingadults are relativelymobile,
(perennial sites) and intermittent
sites is likely to delay recovery,
travellingfrom perennial to more distant intermittent
298
THE
AMERICAN MIDLAND
NATURALIST
131(2)
particularly in summer when high temperatures and low humidity reduce adult longevity
by as much as 50% and mortality rates are high (Jackson, 1988). Before rewetting in 1990,
there was more than 2.5 km of dry channel upstream and 1.2 km downstream of site I,
whereas perennial stream reaches were within 800 m of site II, and, consequently, resilience
was greater at the latter site. However, more information on recovery at sites of varying
flow permanence and distance from perennial sites (i.e., from a source of colonists) is needed
to further develop and test these hypotheses.
Invertebrates cope with intermittency in desert streams via resistance to drying and
resilience after rewetting. Resistance to drying is characterized by tolerance of physicochemical extremes in drying pools, and/or avoidance of desiccation by moving upstream or
flyingamong sites during periods of water loss. However, as few taxa demonstrate resistance
during drying, taxonomic shifts are common as non-resistant species become scarce. This
has important implications for postflood recolonization trajectories for floods occurring
during drought periods. There is virtually no survival after water loss; resilience, rather
than resistance, has been favored among Sycamore Creek and other arid land stream benthos
(Harrison, 1966; Hynes, 1975; Gagneur and Chaoui-Boudghane, 1991). Adaptations favoring recovery by recolonization allow invertebrate populations to persist through both
flooding and drying in these lotic ecosystems where such hydrologic extremes are the rule
rather than the exception.
was supportedby NSF grant BSR 88-18612 to S. G. Fisher. D. L.
Acknowledgments.-Research
Buschman was supportedby a Research Experience forUndergraduatessupplement.A. J. Boulton
receivedan AustralianResearch Council fellowshipand E. H. Stanleywas supportedby an American
Fellowship fromtheAmericanAssociationofUniversityWomen during1991-1992. W. L. Minckley,
J. P. Collins, J. J. Elser and two anonymousreviewersgave many valuable commentson an earlier
draftof this manuscript.John Whitneyprovidedaccess to samplingsites at SycamoreCreek.
LITERATURE CITED
streams,p. 46-60. In: S.
of some California intermittent
ABELL,D. L. 1984. Benthicinvertebrates
streams.Instituteof Ecology PubJain and P. Moyle (eds.). Vernal pools and intermittent
licationNo. 28, Davis, California.
BENZIE,J. A. H. 1984. The colonisationmechanismsof streambenthosin a tropicalriver(Menik
111:171-179.
Ganga: Sri Lanka). Hydrobiologia,
BERRA, T. M., D. M. SEVER AND G. R. ALLEN.
1989.
Gross and histological morphology of the
an
swimbladderand lack of accessoryrespiratorystructuresin Lepidogalaxiassalamandroides,
aestivatingfishfromWesternAustralia. Copeia, 4:850-856.
in two intermittent
BOULTON,A. J. 1989. Over-summeringrefugesof aquatic macroinvertebrates
streamsin centralVictoria. Trans.Roy. Soc. S. Aust.,113:23-34.
streamsin Victoria,Australia. II.
ANDP. S. LAKE. 1992a. The ecologyof two intermittent
Comparisons of faunal compositionbetweenhabitats,rivers,and years. Freshwat.Biol., 27:
99-121.
streamsin Victoria, Australia. III.
. 1992b. The ecology of two intermittent
AND
Temporal changes in faunal composition.Freshwat.Biol., 27:123-138.
C. G. PETERSON, N. B. GRIMM AND S. G. FISHER.
1992a.
Stability of an aquatic macro-
E. H. STANLEY, S. G. FISHER AND P. S. LAKE.
1992b.
Over-summering strategies of
invertebratecommunityin a multi-yearhydrologicdisturbanceregime. Ecology, 72:21922207.
in intermittent
streamsin Australia and Arizona, p. 227-237. In: R. D.
macroinvertebrates
Robarts and M. L. Bothwell (eds.). Aquatic ecosystemsin semi-aridregions:implicationsfor
resourcemanagement.NHRI SymposiumSeries 7, EnvironmentCanada, Saskatoon, Saskatchewan,Canada.
ANDP. J. SUTER. 1986. Ecology of temporarystreams-an Australian perspective,p. 313-
1994
STANLEY ET AL.: RESISTANCE
AND RESILIENCE
299
327. In: P. De Deckker and W. D. Williams (eds.). Limnologyin Australia. CSIRO and
Dr. W. Junk,Melbourne, Australia and Dordrecht,The Netherlands.
H. M. VALETT AND S. G. FISHER. 1992c. Spatial distributionand taxonomiccomposition
of the hyporheosof several Sonoran Desert streams.Arch. Hydrobiol.,125:37-61.
(Limnephilidae)
des Stenophylax
BOUVET, Y. 1977. Adaptationsphysiologiqueset compartementales
aux eaux temporaires,p. 117-119. In: M. I. Crichton (ed.). Proceedings of the Second
InternationalSymposiumon Trichoptera.Dr. W. Junk,The Hague, Netherlands.
1987. Some effectsof the 1982-83 droughton waterquality
CHESSMAN, B. C. AND D. P. ROBINSON.
and invertebratefauna of the lower LaTrobe River,Victoria.Aus.J. Mar. Freshwat.Res., 38:
289-299.
stream.Invest.Indiana Lakes
in an intermittent
CLIFFORD, H. F. 1966. The ecologyof invertebrates
Streams,7:57-98.
1993. Aspects of the hyporheiczone below the terminusof a
COOLING, M. AND A. J. BOULTON.
South Australian arid-zone stream.Aust.J. Mar. Freshwat.Res., 44:411-426.
CROWE, J. H. 1971. Anhydrobiosis:an unsolvedproblem.Am. Nat., 105:563-573.
temporal
C. M. 1988. Comparisonofcommunitystructureamongstreamswithdifferent
DELUCCHI,
flowregimes.Can. J. Zool., 66:579-586.
in temporaryand permanentstreams.Oecologia,
1989. Movementpatternsof invertebrates
78:199-207.
and permanent
1989. Life historypatternsofinsectsin an intermittent
AND B. L. PECKARSKY.
stream.J. N. Am. Benthol.Soc., 8:308-321.
1984. Aquatic insectrespiration,
ERIKSON, C. H., V. H. RESH, S. S. BALLING AND G. A. LAMBERTI.
p. 27-37. In: R. W. Merritt and K. W. Cummins (eds.). An introductionto the aquatic
insectsof North America. 2nd edition.Kendall/Hall, Dubuque, Iowa.
communitiesin a lowland river.
of droughton benthicinvertebrate
EXTENCE, C. A. 1981. The effect
83:217-224.
Hydrobiologia,
1983. Secondary productionand organic matterprocessingby
FISHER, S. G. AND L. J. GRAY.
in a desertstream.Ecology,64:1217-1224.
collectormacroinvertebrates
1982. Temporal successionin a desertstream
, N. B. GRIMM AND D. E. BUSCH.
followingflashflooding.Ecol. Monogr.,52:93-110.
1991. Sur le r6le du milieu hyporheique pendant
GAGNEUR, J. AND C. CHAOUI-BOUDGHANE.
6:77-89.
l'assechementdes oueds de l'Ouest Algerien.Stygologia,
developmentof desertstreammacroinGRAY, L. J. 1980. Recolonizationpathwaysand community
vertebrates.Ph.D. Dissertation,Arizona State University,Tempe, 175 p.
1981. Species compositionand lifehistoriesof aquatic insectsin a lowland Sonoran Desert
stream.Am. Midl. Nat., 106:229-242.
in a
1981. Postfloodrecolonizationpathways of macroinvertebrates
AND S. G. FISHER.
lowland Sonoran Desert stream.Am. Midl. Nat., 106:249-257.
to disturbance
1989. Stabilityofperiphytonand macroinvertebrates
GRIMM, N. B. AND S. G. FISHER.
by flashfloodsin a desertstream.J. N. Am. Benthol.Soc., 8:293-307.
62:
HARRISON, A. D. 1966. Recolonizationof a Rhodesian streamafterdrought.Arch.Hydrobiol.,
405-421.
in a riverin southernGhana. Freshwat.
HYNES, J. D. 1975. Annual cyclesof macro-invertebrates
Biol., 5:71-83.
of selectedadult aquatic insectsfrom
JACKSON, J. K. 1988. Diel emergence,swarmingand longevity
a Sonoran Desert stream.Am. Midl. Nat., 119:344-352.
1959. Destructionand re-establishment
AND C. HECKROTTE.
LARIMORE, R. W., W. F. CHILDERS
of streamfishand invertebratesaffectedby drought.Trans.Am. Fish. Soc., 88:261-285.
1987. Persistenceand stabilityof fish and invertebrate
G. K. AND W. L. MINCKLEY.
MEFFE,
assemblagesin a repeatedlydisturbedSonoran Desert stream.Am. Midl. Nat., 117:177-191.
Oklahoma stream.Am.
W. T. 1966. Upstream movementof crayfishin an intermittent
MOMOT,
Midl. Nat., 75:150-159.
recolonizationof previouslydrychannelsin the Rakaia River. N.
SAGAR, P. M. 1983. Invertebrate
Z. J. Mar. Freshwat.Res., 17:377-386.
300
THE
AMERICAN MIDLAND
NATURALIST
131(2)
R. E. AND R. G. PEARSON. 1987. The macro-invertebrate
communitiesof temporarypools
in an intermittent
streamin tropicalQueensland. Hydrobiologia,
150:45-61.
STANLEY, E. H. 1993. Drying disturbanceand stabilityin a desertstreamecosystem.Ph.D. Dissertation,Arizona State University,Tempe. 371 p.
AND S. G. FISHER.
1992. Intermittency,
disturbance,and stabilityin streamecosystems,p.
271-280. In: R. D. Robarts and M. L. Bothwell (eds.). Aquatic ecosystemsin semi-arid
regions: implicationsfor resource management.NHRI SymposiumSeries 7, Environment
Canada, Saskatoon, Saskatchewan,Canada.
1992. Interactionsbetween dryingand the hyporheiczone, p. 234AND H. M. VALETT.
249. In: P. Firth and S. G. Fisher (eds.). Global climatechange and freshwaterecosystems.
Springer-Verlag,New York, New York.
STEHR, W. C. AND J. W. BRANSON.
1938. An ecological studyof an intermittent
stream.Ecology,
19:294-310.
STRANDINE, E. J. 1941. Effectsof soil moistureand algae on the survivalof a pond snail during
periodsof relativedryness.The Nautilus,54:128-130.
1984. Adaptationsto hazardous seasonal conditions:
TAUBER, M. J., C. A. TAUBER AND S. MASAKI.
dormancy,migration,and polyphenism,p. 149-183. In: C. B. Huffakerand R. L. Rabb
(eds.). Ecological Entomology.J. Wiley and Sons, New York, New York.
TAYLOR, R. C. 1983. Drought-inducedchanges in crayfishpopulationsalong a streamcontinuum.
Am. Midl. NVat.,
110:286-298.
1968. The SycamoreCreek watershed,Maricopa County,
THOMSEN, B. W. AND H. H. SCHUMANN.
Arizona. United StatesGeological SurveyWater Supply Paper 1861. Washington,D.C. 53 p.
1990. Physical and chemicalcharacteristics
of
VALETT, H. M., S. G. FISHER AND E. H. STANLEY.
the hyporheiczone of a Sonoran Desert stream.J. N. Am. Benthol.Soc., 9:201-215.
1974. Nutrientrecyclingand the stabilityof
WEBSTER, J. R., J. B. WAIDE AND B. C. PATTEN.
ecosystems,p. 1-27. In: F. G. Howell, J. B. Gentryand M. H. Smith(eds.). Mineral cycling
in southeasternecosystems.ERDA SymposiumSeries, Washington,D.C.
1980. Evolutionaryand ecological strategiesof
WIGGINS, G. B., R. J. MACKAY AND I. M. SMITH.
animals in annual temporarypools. Arch.Hydrobiol.Suppl., 58:97-206.
D. D. 1977. Movement of benthosduring recolonizationof temporarystreams.Oikos,
WILLIAMS,
29:306-312.
1987. The ecologyof temporarywaters. Croom Helm, London, England. 205 p.
1977. The ecology of temporarystreams.II. General remarkson
AND H. B. N. HYNES.
temporarystreams.Int. Rev. Ges. Hydrobiol.,62:53-61.
N. E. WILLIAMS AND H. B. N. HYNES. 1974. Observationson the lifehistoryand burrow
constructionof the crayfishCambarusfodiens(Cottle) in a temporarystream in southern
Ontario. Can. J. Zool., 52:365-370.
W. D. 1985. Biotic adaptations in temporarylentic waters, with special referenceto
WILLIAMS,
those in semi-aridand arid regions.Hydrobiologia,
125:85-110.
ZAR, J. H. 1984. Biostatisticalanalysis, 2nd edition.Prentice-Hall,New Jersey.718 p.
SMITH,
SUBMITTED 17 AUGUST 1993
ACCEPTED 14 DECEMBER 1993