University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
USGS Northern Prairie Wildlife Research Center
Wildlife Damage Management, Internet Center for
1-1-1988
Determinants of Breeding Distribution of Ducks
Douglas H. Johnson
Northern Prairie Wildlife Research Center
James W. Grier
North Dakota State University
Follow this and additional works at: http://digitalcommons.unl.edu/usgsnpwrc
Part of the Other International and Area Studies Commons
Johnson, Douglas H. and Grier, James W., "Determinants of Breeding Distribution of Ducks" (1988). USGS Northern Prairie Wildlife
Research Center. Paper 242.
http://digitalcommons.unl.edu/usgsnpwrc/242
This Article is brought to you for free and open access by the Wildlife Damage Management, Internet Center for at DigitalCommons@University of
Nebraska - Lincoln. It has been accepted for inclusion in USGS Northern Prairie Wildlife Research Center by an authorized administrator of
DigitalCommons@University of Nebraska - Lincoln.
WILDLIFEMONOGRAPH
(ISSN:0084-0173)
A Publicationof The WildlifeSociety
C t141
;s
°
L
l I
@l
X g r
DETERMINANTS
OF BREEDING
DISTRIBUTIONS
OF DUCKS
by
DOUGLASH. JOHNSONANDJAMESW. GRIER
NO. 100
OCTOBER1988
This article is a U.S. government work, and is not subject to copyright in the United States.
:is
:
>
_>
.
r
.
<<-!A./
t-_I'
FRONTISPIECE. Lssser scaup use more permanent wetlands and breed in more northern locations than most of the common
North American ducks. (U.S. Fish and Wildlife Service photo)
DETERMINANTS
OF BREEDING
DISTRIBUTIONS
OF DUCKS
DOUGLASH. JOHNSON
U.S. Fish and Wildlife Service, Northern Prairie Wildlife Research Center, Jamestown, ND 58402
JAMES W. GRIER
Zoology Department, North Dakota State University, Fargo, ND 58105
Thesettlingof breedinghabitatby migratorywaterfowlis a topicof boththeoretical
andpractical
interest.We usetheresultsof surveysconductedannuallyduring1955-81in majorbreedingareasto examine
thefactorsthataffectthedistributions
of 10 commonNorthAmericanduckspecies.Threepatternsof settling
aredescribed:
homing,opportunistic,
andflexible.Homingis generallymorepronounced
amongspeciesthat
use more stable(morepredictable)wetlands,such as the redhead(Aythya americana), canvasback(A.
valisineria), lesserscaup(A. affinis), mallard(Anas platyrhynchos), gadwall(Anas strepera), and northern
shoveler(Anas clypeata). Opportunistic
settlingis moreprevalentamongspeciesthat use less stable(less
predictable)wetlands,suchas northernpintail(Anas acuta) and blue-wingedteal (Anas discors). Flexible
settlingis exhibitedto variousdegreesby mostspecies.
The 10 speciesareshownto fall alonga naturalordinationreXectingdiSCerent
life historycharacteristics.
Averagevaluesof indicesof r- andK-selectionindicatedthatpintail,mallard,blue-wingedteal,andshoveler
havethe mostfeaturesassociatedwithunstableor unpredictable
environments.
Gadwall,Americanwigeon
(Anas americana), andgreen-winged
teal(Anas crecca) wereintermediate,
andattributes
of thedivingducks
wereassociatedwith the use of stableor predictableenvironments.
Somespecies notablymallard,gadwall,blue-wingedteal,redhead,andcanvasbacktendto fillavailable
breedinghabitatfirstin the centralportionsof theirrange,andsecondlyin peripheralareas.OtherspeciesAmericanwigeon,green-wingedteal,northernshoveler,northernpintail,andlesserscaup fill theirhabitat
in the orderit is encounteredduringspringmigration.
Age andsex classeswithinspeciesvaryin theirsettlingpattern.Someof thisvariationcan be predicted
from the matingsystemsof ducksin which breedingfemales,especiallysuccessfulones, have a greater
investmentin habitatresourcesandare morelikelyto returnto the sameareain subsequentyears.
Abstract:
WILDL.
MONOGR.100, 1-37
WILDLIFEMONOGRAPHS
6
CONTENTS
INTRODUCTION
Acknowledgments
6
7
7
A THEORYOF SETTLING
Homing
8
OpportunisticSettling
8
FlexibleSettling
8
LifeHistoryCharacteristics
9
METHODS
9
Survey Design
9
AerialCounts
11
GroundCounts
11
Pond Counts
11
Computations
12
Analysisof Pond Counts
12
Breeding-season Distribution,WinteringAreas,
and MigrationCorridors
12
Relations of Ducks with Local and Continental
Ponds
12
Relationsto LifeHistoryCharacteristics
13
RESULTS
13
Dynamicsof the WetlandHabitat
13
INTRODUCTION
Breeding Distributions of the Species
Mallard
Gadwall
American Wigeon
Green-vinged Teal
Blue-winged Teal
Northern Shoveler
Northern Pintail
Redhead
CanvasUck
Lesser Scaup
General Patterns
Relations with Life History Characteristics
DISCUSSION
Settling Patterns
Fillingthe Habitat
Relations to Life History Characteristics
Management Applications
LITERATURECITED
13
13
16
16
21
21
24
24
24
27
27
30
30
31
31
32
32
34
35
ing estimatesof the breedingpopulation
Thesettlingofhabitatbymigratory
birds withmeasures
of productivity
(Martinet
withintheirgeographicrangerepresents al. 1979). Becausereproductionvaries
habitatselectionona coarsescale(Johnson widelyamongbreedinggrounds(Hansen
1980),finerthangeographical
range,but andMcKnight
1964,CalverleyandBoag
broaderthanhomerangeandmicrohab- 1977,Derksenand Eldridge1980),unitat.Analysisof settlingbehavioron this derstanding
how ducksdistributethemscaleraisesbothbiologicalquestionsand selvesthroughout
theirrangeshouldpromanagement
issues.
videusefulinformation
forpredicting
the
Onthebiologicalside,thereis a consid- size of fall populations.
At a locallevel,
erablebody of theoryconcerninggeo- the effectivenessof land management
graphicdistribution
of highlymobilean- techniquesto increasenumbersandproimals such as birds.Both the ultimate ductionof ducksdependsontheextentto
factorsthatinfluenceevolutionary
fitness whichducksreturnto the sameareain
in the habitatselectedandthe proximate subsequent
yearsorseekoutnewbreeding
cuesusedbyanimalstoselecthabitats
have locations(Hochbaum
1946).
beenintensively
studied.Lifehistorychar- Ourobjectiveswereto determineecoacteristics
of differentspecieshavebeen logicaland geographiccorrelatesof the
relatedtohabitats
occupiedbythespecies; breedingdistributionsof 10 common
muchof the discussion
on this topichas speciesof NorthAmericanducksand to
beencastin termsof r- and K-selection relatetheseto the life historycharacter(Pianka1974).Distributions
andbasicbi- isticsof the species.Thespeciesaremalologyareparticularly
wellstudiedinducks lard,gadwall,Americanwigeon,green(Johnsgard
1975, Palmer1976, Bellrose wingedteal, blue-wingedteal, northern
1980),thusproviding
opportunities
forde- shoveler,
northern
pintail,allof whichare
tailedanalysis.
dabblers
(TribeAnatini),andcanvasback,
Thenatureof settlingpatternsof ducks redhead,andlesserscaup,alldivers(Tribe
isimportant
towaterfowl
managers
atboth Aythyini).
T}aisanalysisis restricted
to the
continentaland locallevels.To regulate areain NorthAmericaregularly
surveyed
hunting,for example,managerspredict forbreedingducks(Fig.1),anddataused
the sizeof the fall population
by combin- areresultsof thosesurveys.
OF DUCKS
BREEDINGDISTRIBUTIONS
Johnsonand Grier
7
10
Fig. 1. Strata used in breedingwaterfowlsurveys.
We propose3 patternsthat migratory H. Boyd,W. F. Crissey,A. Dzubin,J. L.
birdscoulduseforsettlingbreedinghab- Eldridge,R. R. Koford,G. L. Krapu,J. T.
H. W.Miller,J.D. Nichols,T.
itat.We thendescribethe methodsused Lokemoen,
andA. B.
to gatherandanalyzethe data.Distribu- D. Nudds,G. L. Nuechterlein,
numerous
offered
Miller
and
R.
M.
described
are
Sargeant.
tionsof breedingducks
themanuscript.
thatimproved
relatedto winteringareasand migration suggestions
corridors,local and continentalwetland
conditions,and homingand pioneering A THEORYOF SE1TLING
tendenciesof each species.We also debirds,by virtueof theirmoMigratory
amongthespeciesandhow
scriberelations
ourfindingsfitwiththespecies'lifehistory bility, have many optionsfor selecting
We concludewith some habitatin whichto breed(Fretwelland
characteristics.
Lucas 1969, Gauthreaux1980, Bailey
for management.
implications
grateful
are
1981).The qualityof a habitathas2 asWe
Acknotvledgments.
andothersat theOfflce pects-the ultimatequalityandthe proxto R.S. Pospahala
U.S.Fish imatecuesby whichthe birdschooseit
BirdManagement,
of Migratory
survey (Immelmann
1973).
andWildlifeService,forproviding
Ultimatefactorsare thosethat affect
datausedin the analyses,and to A. D.
sucAftonandR. A. Wishartfor useof their long-termsurvivaland reproductive
unpublishedinformationon homingof cess.Optimalhabitatfor survivalof the
lesserscaupand Americanwigeon,re- adultmay differfromthe optimumfor
forreproducrequirements
spectively.We appreciatethe valuable reproduction;
commentson variousdraftsof thisreport tiontendto be morestringent.Not only
but
F. C. Bellrose, mustthebirdbe safefrompredation,
madeby M. G. Anderson,
8
WILI:)LIFEMONOGRAPHS
so musttheeggsandducklings,
life stages OpportunisticSettling
thatarehighlyvulnerable.
Foodsupplies
mustincludethe minimumforadultsur- Thesecondpatternis fora birdto settle
alongits mivival,aswellasadequatecalcium,protein, in the firstsiteencountered
gration
path
that
appears
to
offer
thereqandenergyforeggproduction
(Holmand
uisites
for
survival
and
breeding.
This
patScott1954,Krapu1979),andnutrients
for
migration'>
the growthanddevelopment
of young. ternis akinto the "facultative
Proximate
cuesusedby ducksto iden- of PulliamandParker(1979)andwould
if habitatconditionsare
tify suitablebreedingareasareunknown, be appropriate
unpredictable
from
year to year.Morebutarethoughtto be indirectandrelated
settlingminimizes
any
stronglyto wetlandfeatures.Empirical over,opportunistic
costs.Dispersal,
suchas caused
evidencecomesfromstudiesinwhichwet- migration
settling,is an important
landswerecreatedormarkedly
improved, by opportunistic
means
of
adjusting
to majorchangesin
oftenleadingtolargeincreases
inthenumber of breedingducks(e.g., Hochbaum birds'habitatsandis especiallyimportant
habitats(Gau1946,Hochbaumand Bossenmaier
1972, in spatiallyheterogeneous
of
McKnight
1974).Although
duckstypically threaux1980,1982).Onemanifestation
opportunism
is
frequent
pioneering,
the
usea varietyof wetlands
duringthebreedof birdsintohabitatsthatwere
ing period(Dzubin1969, Dwyeret al. movement
1979),manyof whichdo not persistfor unavailableor not occupiedduringthe
hasbeendocutheentirebreedingseason,the conditions previousyear.Pioneering
mented
by
anecdotes
attributing
increases
of wetlandsin springoffergoodcuesto
in
duck
numbers
in
a
local
area
as
a rapid
pondconditions
inthesubsequent
summer
(LeitchandKaminski
1985).Thecustom- responseto improvedhabitatand by
arymeasureof wetlandquantityin exten- large-scalesurveysthat showa redistrisive waterfowlstudiesis the numberof butionof ducksfrom1 yearto the next
wetlandbasinsthatcontainwaterduring when wetlandconditionsin partof the
somesurveyperiod,typicallyMay(Cris- areachangemarkedly.
sey 1969).
We hypothesize3 patternsof settling FlexibleSettling
behaviorand the ecologicaland species
characteristics
associatedwith each pat- A mixedpatternis for a birdto home
tern:(1)homing,(2)opportunistic
settling, to the areaused the previousyear,but
and (3) flexiblesettling.It couldbe ad- moveon(i.e.,becomeopportunistic)
if that
vantageous,
however,for differentage or habitatis not suitable.Droughtdisplacesex classeswithina speciesto followdif- ment (theoverflightof normalbreeding
ferentpatterns.
areas)is includedhere.Thepatternissimilarto the <'iElexible
homing"to wintering
areas
discussed
by
Bellrose
andCrompton
Homing
(1970)andcanbeviewedasa compromise
Onepatternis alwaysto home,thatis, betweenhomingand opportunistic
setfor adultsto returnto the breedingarea tling.Thehabitatchosenaftermovingon
usedthe previousyearand for yearlings mayenhancesurvivalof the adultbutreto returnto theirnatalarea.Suchphilo- ducereproductive
performance.
Droughtpatryhasbeentermed"migrational
hom- displacedbirdsare reportedto havelow
ing" (Sowls1955).Homingcan be pre- reproductive
successs(Hansenand Mcdicted if the temporal variabilityof Knight1964,Murdy1966,Calverleyand
resources
is low,butwouldnotbe advan- Boag1977,Derksenand Eldridge1980),
tageousotherwise(Greenwood
and Har- and wetlandsin areas frequentedby
vey 1982).Simplyif the birds'essentials drought-displaced
nonbreeding
birdsare
areprovidedeachyear,thereis noadvan- typicallylessproductivethanthoseused
tageto switchingbreedingareas
by breedingbirds(Murdy1966).
BREEDINGDISTRIBUTIONS
OF DUCKS-Johnson
and Grier
9
In NorthAmericamost drought-dis-selects.The theoryof settlingEaresented
responses
placedducksarethoughtto movenorth- heresuggeststhatopportunistic
with
ward or northwestward(Hansenand to habitatchangewill be associated
McKnight
1964,Crissey1969Smith1970), the useof variablehabitatsandwithrelspecies,andthathoming
although
Stoudt(1971)suggested
thatblue- ativelyr-selected
wingedteal may movesouthward.
Evi- will be associatedwith the use of stable
dencefordrought
displacement
arisesfrom habitats(e.g.,Wiens1976)and with relspecies.Wehypothesize
thefollowing2 sources:
(1)localstudiesin ativelyK-selected
northern
areasthatshowgreaterdensities that a given settlingpatternor selected
of ducksSandsometimesthe presenceof breedinghabitatis withina suiteof life
speciesnot usuallyfoundthere,in years historycharacteristics
thatcanbe regardof droughtin the prairieand parkland ed as relativelyr-selectedor K-selectedS
thatotherkindsof se(Hansen
1960}HansenandMcKnight
1964> whilerecognizing
Derksenand Eldridge1980);and (2) ex- lectioncouldhavecausedthesesuitesto
tensivesurveysthatshowgreaterpropor- covary(Stearns1977).If we are correct,
tionsof ducksin northernareasthanin then speciesof duckscouldbe arrayed
prairieand parklandduringyearsof se- alongan axisthatreflectssettlingpattern
vere prairiedrought(Hansenand Mc- andhabitatselection;
thisaxiswouldcorKnight1964,Crissey1969,Smith1970). respondto an axisof r- and K-selection.
Severalauthors(Patt&rson
1979 Bailey
1981,VickeryandNudds1984)havedone
Life History Characteristics
relatedcomparisons,
mostlyattempting
to
Ecologistslong have recognizedthat placeduckspeciesalonga continuumof
manylife historycharacteristics
areasso- r- andK-selection.
ciatedwithoneanother,
e.g.,longevityand
fecundity.Effortshavebeenmadeto relatesuitesof characteristics
to featuresof METHODS
the habitatusedby the animals,specifi- Survey Design
callycontrasting
highlyvariable(andunpredictable)
environments
withthosemore Eachyearducksandpondsarecounted
stable(andpredictable)
(e.g. Pianka1974). in the majorNorthAmericanbreeding
Comparisons
oftenhavebeencastinterms areas.The surveydesignand datagathof r- andK-selection
(MacArthur
andWil- eringtechniques
weredescribedin detail
son 1967),commemorating
the parame- by D. C. Bowden(Reviewandevaluation
tersof thelogisticequationof the growth of Maywaterfowlbreedinggroundsurof populations
subjectedto highlevelsof vey, unpubl.rep., U.S. Fish and Wlldl.
density-independent
(r-selectedpopula- Serv.,Patuxent,
Laurel,Md*1973)Anontions) or density-dependent
(K-selected ymous(1977)andMartinet al. (1979).
populations)
mortality
(Stearns
1977,Boyce The regionfrom which samplesare
1984).Theintroduction
of theseparame- drawnconsistsof about3.03 millionkmS
tershasbroughtmuchcontroversy
to the (1.17millionsq miles)in thenorth-central
discussion
of covaryinglife historychar- UnitedStates,the prairieprovincesand
acteristics.Pendinga resolutionof this NorthwestTerritoriesof Canada and
controversywe will use the expressions,Alaska(Fig. 1). About78-84%of North
r-selection
andK-selection,
torefertQ pOp- Americanmallardsbreedwithinthe surulations
thatareinfluenced
relatively
more veyedarea(Pospahala
et al. 1974;R. S.
by factorsthatoperatein a density-inde- Pospahala,
unpubl.memorandum
1985).
pendentmannerandpopulations
morein- Becausetheother9 specieshavebreeding
flueneedby density-dependent
factors,re- rangesmorerestrictedto thatarea(Bellspectively.
rose 1980) even greaterpercentagesof
Thesettlingpatternof a speciesmaybe theircontinental
populations
probably
nest
relatedto the stabilityof the habitatit in the surveyedarea.
WILDLIFEMONOGRAPHS
10
Table 1. Strata used in May waterfowl surveys: name, area, number of transects, and area sampled, 195541.
-
Name
Stratum
2
3
4
s
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Total
Kenai-Susitna
Nelchina
Tanana-Kuskokwim
Yukon Flats
Innoko River
KoyukukRiver
Copper River Delta
BristolBay
Yukon Delta
Seward Peninsula
Kotzebue
Old Crow Flats
Mackenzie River Delta
Northwest TerritoriesForest Tundra
Northwest TerritoriesForest Tundra
Slave River Parklands
Northwest TerritoriesClosed Forest and Forest Tundra
Northwest TerritoriesPre-CambrianEdge
North Alberta Parklandsand Closed Forest
AthabascaRiver Delta
Northern SaskatchewanMixed Forest
Northern SaskatchewanMixed Forest
Northern SaskatchewanPre-CambrianShield and Mixed
Forest
ManitobaPre-CambrianShield, Muskeg, and Mixed Forest
SaskatchewanRiver Delta
Alberta Parklands
Alberta Prairie
Alberta Prairie
Alberta Prairie
SouthernSaskatchewanMixed Forest (west)
SouthernSaskatchewanMixed Forest (east)
SaskatchewanPrairie
Cypress Hills
SaskatchewanParklands
SaskatchewanParklands
ManitobaInterlake Region
ManitobaInterlake Region
ManitobaParklands
ManitobaParklands
ManitobaParklands
MontanaNorth River
MontanaSouth River
North Dakota West River
South Dakota West River
Drift Prairie and MissouriCoteau (ND)
Drift Prairie and MissouriCoteau (ND)
Red River Valley (ND)
MissouriCoteau, James River Valley, and Prairie Coteau
(SD)
South and Eastern South Dakota
-
a
nsectsthan indicatedwere surveyedin some years.
Fewer trar
Area
(sq miles)
No. of
transectsa
Area
sampled
(sq miles)
2,200
3,900
6
10
40
52
9,300
18
132
0,800
12
80
3,400
7
44
4,100
10
80
400
9,900
1l
26,600
3,850
1
20
92
8
4
260
24
5,350
7
48
1,970
4,935
79,217
3
4
8
36
36
364
50,462
57,821
2
4
108
94.5
69,033
,733
5
3
297
27
63,861
8
688.5
1,625
4
44,922
1l
31.5
382.5
20,278
6
166.5
68,320
6
297
83,794
6,816
8
6
490.5
117
26,448
,724
2,890
3,235
18,570
21,086
37,911
,345
3,164
6
265.5
3
3
4
144
135
171
4
14
162
135
571.5
6
5
90
175.5
126
5
9,044
6
5,500
3
58.5
6,485
4
130.5
5,655
3
54
6,552
4,536
32,902
5
4
7
121.5
67.5
504
40,755
19,835
7
5
364.5
175.5
27,299
26,625
4,238
5
7
8
216
310.5
270
7,821
24,587
6
9
45
315
15,830
1,168,624
11
171
8,787.5
BREEDING
DISTRIBUTIONS
OF DUCKS Johnsonand Grier
ll
2. Years in which waterfowl breeding-pair surveys and
Theregionis dividedinto49 strata(Ta- Table
pond surveys were initiated, by stratum.
ble1),ranginginsizefrom1,036to424,400
km2(400-163,861sq miles).Stratification
Strata
Breeding pairs
Ponds
is basedbothon politicalboundaries
and
1-11
1957
Not done
on geographical
differencesin duckden12-20
1956
Not done
sities(Stewartet al. 1958).Anderson
and
21-25
1955
Not done
26-40
1955
1955
Henny(1972)brieflydescribed
majorwa41-42
1965
1965
terfowlhabitatsin thesurveyedarea.Wa43,45 - 47
1958
1958
terfowlcountswithinthe variousstrata
44,48-49
1959
1959
wereinitiatedin yearsrangingfrom1955
to 1965(Table2). We useddatathrough Data missing in 1963, 1965, and 1966 for stratum 43.
1981.
Withineachstratum,a systematic
sam- theaerialsurvey.Effortsaremadeto cenpleof transectlineswasdrawn.Thestart- susall duckswithinthe boundaries
of the
inglinewasselectedrandomly(Martin
et sampleunit.
al. 1979).Transects
are400-m(0.25-mile)
ratioof birdsof eachspeciescountwideandof variablelength.Aerialcounts edThe
on
the
groundto the numbercounted
of ducksand pondsare conductedeach fromthe air,
takenoverseveralsegments
yearalongeachtransect,
anda subsample withina group
of strata,istermedthe"visof portions
of transects
is canvassed
onthe ibilityrate,"although
itsreciprocal
might
groundforducks.
moreaptlybearthatname.Visibility
rates
are calculatedby speciesand year for
AerialCounts
groupsof strata.Forthe northernstrata,
groundsurveysarenotconducted,
Fixed-wingaircraftareusedin the ae- where
visibility
ratesaretakenas the long-term
rialcountsandare flownalongthe tran- averagefrom
stratawheretheyaredetersectsata heightof 30-46m (100-150feet) mined.
abovegroundandat a groundspeednot
toexceed169km/hr(105miles/hr).Dates
of thesurvey,in May,arestandardized
to Pond Counts
conformto geographicaldifferencesin
phenological
events.Time of the survey DuringMayaerialsurveys,a countof
andflightconditions
arealsoprescribed. pondbasinscontainingwateris madein
Theaircraftpilotandanotherobserver strata26-49.Includedin thetallyarenatrecordall identifiable
duckswithintran- uralwetlands
of typesIII,IV,andV (Shaw
sect boundaries
(200m = 0.125mile on andFredine1956),whicharebasinsthat
eithersideof theflightline).Unidentified areinundated
seasonally,
semipermanentducksarenot recorded.Factorsinfluenc- ly, or permanently,
respectively.
Streams
ing aerialcountsweredescribedby Mar- andriversthatmeanderacrossthetransect
tinsonandKaczynski
(1967).
arecountedeachtimetheyenterthetransect.Alsoincludedin the pondcountare
man-madewetlandssuchas stockdams,
GroundCounts
dugouts,and largeditchesthatmaintain
To estimatethe proportionof ducks waterintosummer.Waterareasexcluded
presentthatareseenandidentifiedby the fromthe countare temporarywetlands
aerialsurveycrew,groundcountsarecon- and sheetwater(expectedto persist93
ductedon certainsegmentsof sometran- weeks),roadsideor borrowditches,and
sects.Groundcountsarenotmadein the muskegareas.Waterbodiesin the northnorthern
areas(strata1-25)becauseof in- ernstrata(1-25)aremorepermanent
than
accessibility.
Crewsof 2-4 biologists
make thosein prairieand parklandstrataand,
groundsurveys,preferablythe day after therefore,
arenotcounted.Pondcountsfor
a
12
WILDLIFEMONOGRAPHS
strata26-49wereinitiatedatvarioustimes Breeding-season Distribution,
duringthe 1955-65period(Table2).
Wintering Areas, and
Migration Corridors
Computations
of ducks
distributions
Breeding-season
theaverage
Countsof ducks(byspecies)andareas wereexaminedby computing
surveyedaresummedacrossall transects adjusteddensityof eachspeciesfor each
acrossall years
withina stratum,and theirratio,multi- of the 49 strata(averaged
Densitieswere
surveyed).
were
strata
that
pliedby the visibilityrate,is termedthe
accordingto
maps
on
plotted
and
ranked
muldensity,
Adjusted
adjusteddensity.
tipliedby areaof the stratum,yieldsthe thequartilein whichtheyfell:highest12,
Thesum secondhighest12, thirdhighest12, and
ina stratum.
population
estimated
of thesevaluesacrossall stratais the es- lowest13. Plottingaccordingto quartiles
of distributions
thecomparison
timatedpopulationwithinthe surveyed facilitated
area.A similarmethodis usedto deter- amongspeciesdespitedifferencesin avminepondcount,exceptthatnocorrection eragedensity.
The directionfrom which ducksapforvisibilityis made.
groundwas
breeding
proachtheirpotential
andmiarea
wintering
by
their
detemined
Analysis of Pond Counts
grationroute.The pathstheytraverseon
Pondcountsprovidethe onlyindexto theirspringtripsconvergewithinrather
(Bellrose
corridors
migration
breedinghabitatavailableformostof the well-defined
or
north-south
generally
are
Most
1980).
surveyedarea.Countsin a stratumfor a
but
in
orientation,
northwest-southeast
particularyear are dividedby the area
arefairlycommon.
surveyedwithinthatstratumto produce lateralmovements
anestimateof ponddensity.Ponddensities Patternsof use of migrationcorridors
from
variedby yearandevenmoreso by stra- for the 10 speciesweredetermined
Bellrose
by
summarized
information
tum.
of
Becausepondcountsin differentstrata (1980).We estimatedthe proportions
speciesthatreturn
of a particular
wereinitiatedat varioustimesduringthe members
1955-65 period,the data were unbal- to theirmajorbreedinggroundsthrough
encorridors,
orsoutheastern
anced,as not all strataweresurveyedin (a)southern
(b)
westand
Dakotas,
the
through
tering
by
allyears.Asa result,theusualaverages
corridors,entering
To ern or southwestern
yearorby stratummaybe misleading.
Montanaor A1fromthisim- througha northwestern
resulting
problems
overcome
of migration
importance
The
portal.
berta
LeastSquares(LS)
balance,we calculated
theyaffect
is
that
analysis
our
to
corridors
means,in additionto the usualmeans.LS
be
distributed
may
population
the
the
way
linear
meansare basedon an assumed
pondden- relativeto thehabitat;forexample,a pinmodel,in thiscaserepresenting
to
of effects tail migratingdirectlyfromCalifornia
sityas an additivecombination
wetby
be
affected
to
likely
is
less
Alberta
due to stratumandto year.LSmeansby
inSouthDakotathanwould
byadd- landconditions
year,asanexample,arecalculated
ing the interceptof the linearequation, one returningfrom a winteringareain
the effect for the desiredyear,and the Texas.
effects.Thismethod
averageof allstratum
is employedby the SASGLMprocedure Relations of Ducks with Local
Inc.1982).Thelinearmod- and Continental Ponds
(SASInstitute,
el of ponddensities,with 26 yearterms
and23 stratumterms,produceda coeffl- To determinehow localwetlandconR2= 0.75 for the ditionsaffectedthe distribution
of ducks,
cientof determination
coefficient
we calculatedthe correlation
601observations.
OF DUCKS Johnsonand Grier
DISTRIBUTIONS
BREEDING
13
Therelativeextent
attribute).
betweenthe densityof each speciesand aK-selected
in the responseto habitat
thedensityof pondsamongyearswithin of opportunism
Thiscoeff:icientchangewas indexedby the mediancoreachtransectandstratum.
and pond relationbetweenduckdensitiesandpond
numbers
indicateshow duck
highvaluesheresuggestan advariedtogetherfromyeartoyear densities;
numbers
environments
withina transect.For analysis,we aver- aptationto unpredictable
an r-selectedattribute).
overall tran- (hypothetically,
coefflcients
agedcorrelation
arereme- We predictthatthese2 measures
sectswithina stratum.Wecalculated
other.
each
to
lated
where
means,
weighted
and
means,
dians,
the weightwas the numberof yearson
wasbased. RESULTS
coefficient
whichthecorrelation
All 3 measuresperformedsimilarly,and
areusedinthefollowinganalyses. Dynamics of the Wetland Habitat
medians
maydependnotonly Averageponddensitiesvariedwidely
Duckdistributions
butalsooncononlocalhabitatconditions
the strata.Highestdensitieswere
ditionselsewherein the breedingrange, among
Manitobaand
in southwestern
observed
especiallyduringdrought.To assessthis southeastern
and lowest
Saskatchewan
effect,we calculateda measureof total densitiesalongthe easternand western
area(strata edgesof theDakotasandin Montana
pondsfor the pond-surveyed
(Fig.
wasthe
26-49). Foranyyear,thismeasure
2).
ponddensityfor a stratummultipliedby
amongyearsin pondcounts
Variation
andthensummed was
theareaof thatstratum,
asthe prairieandparkland
expected,
overall strata.We usedonlythe datafor areais wellknown
droughts
forrecurring
1966-81, the periodduringwhichponds
means
LS
periods.
wet
intervening
and
were countedthroughoutstrata26-49. were lowerthan ordinarymeans,
espebetweenthe den- ciallyinthefirstfewyears(Fig.3).Surveys
coefflcients
Correlation
sityofeachspeciesandthetotalpondcount werebegunduringa verywetperiod;pond
werecalculatedfor eachspeciesoverall densities
neveragainreachedthenumbers
yearswithina transectandstratum.The attained
1955.Aftera declinefrom1955
in
medianof thesecoefflcientsacrosstran- to 1959,densities
of pondshaveoscillated
sectswithina stratumwasusedas a mea- arounda mean of
about4/km2 (10/sq
varied
sureof howa speciesinthatstratum
mile).
in relationto totalponds.
Relations to Life History
Characteristics
Breeding Distributions
of the Species
Mallard. Mallardsattainedhighest
We attemptedto measureseverallife
foreachspecies.We densitiesin prairieand parklandof the
historycharacteristics
andin theCopprairieprovinces
of southern
thestabilityorpredictability
quantified
Riverdeltas(Fig.
eachspecies'habitatby developingan in- perRiverandAthabasca
area,
of basinwetlands 4A).Withintheprairieandparkland
dexto the permanence
usedby the speciesin an extensivestudy densitiesdecreasedwith distancefrom
by Stewartand Kantrud(1973)in North strataof highestdensities.
Mostmallardswinterin the south-cenDakota.Croplandpondswere scored2,
seasonalpondswerescored3, and semi- tralUnitedStates.We estimatefromin(1980)thatabout70%
inBellrose
permanentponds and reservoirswere formation
returnin thespring
mallards
mulwas
wintering
of
scored4. Theweightedaverage
tipliedby -0.1; high negativevaluesof throughthesouthernentry,or portal(the
versusabout30%fromCaliforthe resultingindexthussuggestconsider- Dakotas),
throughthe
ableuseof stablehabitats(hypothetically,niaandthePacificNorthwest
14
WILDLIFEMONOGRAPHS
Ea hi9hest
Fig. 2.
quartiie
a3 third quartile
;1 lowest quartile
Pond density by stratum, 1955-81.
westernportal.Thusmostmallardsini- of mallardsdisplacedby droughtin the
tiallyareexposedto wetlandconditions
in southern
prairie-parkland
breedingrange.
thesouthern
portionof thesurveyedarea. Femalemallards,especiallythosethat
Mallard
densitiesvariedwithlocalpond weresuccessfulin hatchinga clutchthe
counts,the mediancorrelation
coefficient
being0.417(Table3).Highestcorrelations 10- |
*-*---* CUSTOMARY
ESTIMATES
werein AlbertaprairieandSaskatchewan
LEASTSQUARES
parkland,
butconsistently
highvalueswere
calculatedforthe southernportionof the
prairieprovinces,where mallardswere
verycommon,andin theDakotas(except
the2 easternmost
strata),wheretheywere
notas common(Fig.4B).
Mallarddensitiescovariedpositively
withtotalpondsin muchof theirprimary= .00
¢t t:<Z!7
range(Fig.4C).Correlations
tendedto be
negativein manystratain Alaskaandthe
Northwest
Territories,
aswellasin central
1955
1960
1965
1970
1975
1980
1985
Manitoba
and east-central
Saskatchewan,
YEAR
suggesting
thatthoseareasaredestinations Fig. 3. Pond density (ponds/km2) by year.
\
MALLARD
B
highestcsarti le .
E
sec
quartite []
third quartile
median r
m
median r
-g
m
lowesttarti te
and local pond counts, 1955-81;
between species density
stratum, 1955-81; (B) correlation
by
density
species
Fig. 4. (A) Average
1955-81.
density and total pond counts,
WILDLIFEMONOGRAPHS
16 3. Medians and interquartileranges of correlation coefTable
ficients between duck densities and local pond densities, by
species, 1955-81.
wallsarrivingin springare firstexposed
to wetlandconditionsnearthe southern
portal.
of gadwalls
werenotasclosely
Numbers
Correlationswith pond densities
aswere
local
pond
numbers
correlated
with
Interquartile
range
Median
coefflcients
mallards
(Table3).Correlation
0.159
0.417
Mallard
werehighestin the centralDakotas(Fig.
0.205
0.228
Gadwall
5B),andfairlyhighin moststratawhere
0.208
0.143
American wigeon
gadwallswerecommon.
0.181
0.199
Green-winged teal
posGadwalldensitieswereassociated
0.172
0.390
Blue-winged teal
0.200
0.388
Northern shoveler
itivelywithtotalpondsin only5 stratain
0.157
Northern pintail
0.547
andin eastprairieprovinces
thesouthern
0.220
0.154
Redhead
Negative
cor(Fig.
5C).
ern
North
Dakota
0.191
0.192
Canvasback
relations
werefoundinmanystrataincen0.182
0.145
Scaup
tralCanadaas well as in westernSouth
Dakota,suggestingthat these strataregadwalls.
previousyear,homeat fairlyhigh rates ceivedrought-displaced
(Table4).Maleshomeatmuchlowerrates, Gadwalls,
breedespeciallysuccessfully
becausepairingoccursduring ingfemales,havefairlystronghomingtenpresumably
andthefemale dencies(Table4).Often,however,theydo
winterorspringmigration,
leadsthe male to the breedingground nothomeuntiltheirsecondyear,the age
at whichmanygadwallsfirstbreed(J.T.
1965).
(McKinney
We concludethatmallardstendto re- Lokemoen,H. F. Duebbert,and D. E.
states Sharp,Reproductive
strategiesof prairie
areasinthenorthern
turntobreeding
teal,
gadwalls,andblue-winged
Whenwet- mallards,
prairieprovinces.
andsouthern
U.S.FishandWildl.
in thesestrataarenotfa- unpubl.manuscript,
landconditions
N.D.).
vorable,moremallardsthan usualshow Serv.,Jamestown,
strata.If
We concludethat mostsuccessfulfeup in northernandnortheastern
returntoareasusedthepresuccessfullybreedingfemale mallards malegadwalls
homeat a high rate,dispersingfemales viousyear. Younggadwallsand unsucto
likelysettlein response
likelyaremostlyyearlingbirdsor unsuc- cessfulbreeders
range,
intheirprimary
with wetlandconditions
Becausecorrelations
cessfulbreeders.
pondsarehigherin the primaryrangeof especiallyin the southernpart,judging
coefflcients
the mallardthanin the southernportion, fromthepatternof correlation
mal- betweengadwalldensitiesand wetland
wesuggestthatnorthward-migrating
lardsdo not fill the habitatas they en- counts.Whenmanypondsaredryin the
arecountedincencounterit fromthe south,butinsteadfill prairies,
moregadwalls
the coreareafirst.
tralandnorthernCanada.
Gadwall.- Thisspecieshada primary American Wigeon. This species
of themallard's, reachedhigh densitiesin the southern
rangecenteredsouthward
butwasoftenabundant
andwasmostcommonin the centralDa- prairieprovinces
kotasand the southernprairieprovinces inthenorthaswell(Fig.6A).Wigeonwere
southof Canada.Withinprai(Fig.5A).The gadwallwaslesscommon uncommon
habitats,densitiessteadthanmallardsin the northernstrata.As rieandparkland
with the mallard,highestdensitystrata ilydeclinedawayfromtheareaof highest
by stratawithmoderate density,muchlike the mallardandgadweresurrounded
thesouthdensity,anddensitieswerelowerfarther wall.Wigeonwinterthroughout
ernUnitedStates,alongtheYucatanPenaway.
and insulaof Mexico,and in the Caribbean.
wintermostlyinLouisiana
Gadwalls
usingthesouthof springarrivals
Texas. Perhaps90% of spring-arrivingNumbers
probably
slightlyoutnumber
gadwallspassthroughthe southernportal erncorridors
andonly10%comethroughwesternMon- thoseenteringfromthe southwest(Bellfewmaymigrate
tanaorAlberta(Bellrose1980).Mostgad- rose1980).Anadditional
S.
pecles
.
OF DUCKS
BREEDINGDISTRIBUTIONS
Table 4.
Johnsonand Grier
17
Reported return rates for 10 species of ducks, by age and sex,a 1955-81.
Yearling
Adult
Male
Female
Species
Mallard
n
%
13+
%
n
15
42
46
58
60c
24
113
26
119
37
29
16
52
66C
90
Male
Female
%
Sb
5+
14
31
19
68
Study
n
%
n
20
Ob
13
140
14
48
Sowls (1955)
Coulter and Miller (1968)
Doty and Lee (1974)
Bishop et al. (1978)
Lokemoen et al. (unpubl.
manuscript)d
Lokemoen et al. (unpubl.
manuscript)d
Gadwall
12
8
21
164
Ob
9
Sowls (1955)
Gates (1962)
Lokemoen et al. (unpubl.
manuscript)d
40e
Lokemoen et al. (unpubl.
101
manuscript)d
American
wigeon
Green-winged
teal
Blue-winged
teal
41
54
9
236
38
21
9
11
14
o
Canvasback
Lesser
scaup
Ob
58
16
143
20
39
44
70
70
23
74
75
76
12
75
20
66
330
58
30
Ob
1+
Blohm (1978, 1979)
R. A. Wishart (pers. commun.)
Ob
19
Sowls
Ob
12
SOW1S
200
226
( 1955)
McHenry (1971)
Lokemoen et al. (unpubl.
manuscript)d
19
42
15
42
6
2
0
No information
No information
5
Northern
shoveler
Northern
pintail
Redhead
28
3
7
33
11
19
52
10
34
3
351
9
8b
3
13b
12
116
1 15
15
92
24
27
50
101
626
76
1
134
2b
132
( 1955)
Poston (1969)
Sowls (1955)
206
557
91
Johnson (1978)
Alliston (1979)
Bellrose (1980)
Anderson (1985)
Trauger (1971)
A. D. Afton (pers. commun.)
1
0
4
12
49
a Reportedratesare minimalestimatesof homingrates,as they do not accountfor mortalitybetweenseasonsor for birdsthat may have homed
but were not seen.
b Captive-reared.
c Successfullynested previousyear.
dJ. T. Lokemoen,H. F. Duebbert,and D. E. Sharp, Reproductivestrategiesof prairie mallards,gadwalls,and blue-wingedteal, unpubl.
manuscript,U.S. Fish and Wildl. Serv.,Jamestown,N.D.
e Did not successfullynest previousyear.
di- stratawith high densitiesof the species,
to breedinggroundsin the parklands
locations
intheMid- exceptthat wigeonwere commonin 3
rectlyfromwintering
dle AtlanticStates(Bellrose1980).
parklandstratain Albertaand SaskatchWigeondensitiesgenerallydidnotcor- ewan,butcorrelations
therewerelow.
relatecloselywithlocalponddensities(Ta- Wigeondensitiescorrelatedpositively
between withtotalpondsin manystrataof thecenble 3). The highestcorrelations
wigeonsandlocalpondswerein western tralportionof its range(Fig. 6C). They
part werenegativelyassociated
SouthDakotaandinthesouthwestern
in manystrata,
(Fig.6B).Thispat- particularly
of theprairieprovinces
thoseon the easternedge of
ternis consistent
withthelargemigration its rangeandin the north.
of wigeon throughthe westernportal. The few appreciablecorrelations
beHighercorrelationcoefficientsbetween tweenwigeonnumbers
andpondswerein
wigeonand pondswere associatedwith the southwestern
portionof theirrange,
GADWALL
3second
quartile
1
lowest quartile
>0
40 <0 60
Fig. 5. (A) Average species density by stratum,195541; (B) correlationbetween species density and local pond counts, 1955-81;
densityand total pond counts11955-81.
5
m
second quartile
[2
lowest guartile
by stratum, 1955-81;
Fig. 6. (A) Average species density
density and total pond counts, 1955-81.
: lICAN WIGEON
E
>0,40,<0.60
g
<0.20
a
density and local pond counts, 1955-81;
(B) correlation between species
GREEN-WINGEDTEAL
2
3
hi ghest quart i le
2
th i rd quart; l e
ssond quarti le
C1 lowest warti le
g
>o bo - -
m
median
r
XQ40, C0.60
median r
Smedian
<0.20
]
r
g
Fig. 7. (A) Average species density by stratum,1955-81; (B) correlationbetween species density and local pond counts, 1955-81
density and total pond counts, 1955-81.
OF DUCKS
BREEDINGDISTRIBUTIONS
Johnsonand Grier
21
mostlysouthof the areaof highestden- wingedteal moveto northernstratareto the south.
werehigh, gardlessof pondconditions
sities.Whentotalpondnumbers
Blue-winged Teal.-Blue-winged teal
wigeondensitiesweredepressednotonly
muchlikethatof gadin manystratain the north,but alsoin hada distribution
severalalongthe easternedge,suggesting wall,withhighestdensitiesmostlyin the
and prairieand parklandof Canadaand the
bothnorthward
droughtdisplacement
in
eastward.The limitedevidencesuggests Dakotas(Fig.8A),andwasuncommon
strata.
fairlystronghomingby thespecies(Table thefar-northern
tealwinterin Central
4). Wesuggestthatwigeonarriveontheir Mostblue-winged
ModSouthAmerica.
ornorthern
breedinggroundsfromthe southandthe America
manyremainin stratainitially est numbersare foundin Mexicoand in
southwest;
if pondconditionsthereare the GulfCoastStates(Bellrose1980).A1encountered
(thusaccountingforthe high thougha few mayenterthe majorbreedsatisfactory
withpondsthere);otherwise, ing areadirectlyfromthe east,almostall
correlations
returnin the springthroughthe southern
theyoverflyto the northandeast.
in
Green-upingedTeal. Thisspecieshad portalandarefirstexposedto wetlands
centeredfarthernorththan the Dakotas.
a distribution
of thisspeciesvariedmarkedly
Densities
the otherdabblingducks.Only4 of the
stratawith highestdensitieswere in the withlocalpondcounts(Table3). Consiswerefoundinmost
tentlyhighcorrelations
southernprairieprovinces(Fig.7A).
tealwinterovera broad stratawherethespecieswascommon(Fig.
Green-winged
wereinseveralstrataalong
UnitedStatesandin 8B).Exceptions
rangein thesouthern
andeasternedgesof thecore
Mexico.About75Soof winteringbirdsre- thenorthern
ofblue-winged
turntothemajorbreedinggroundsviathe areathathadlargenumbers
Mostof therestwinter teal,but teal numberswerenot strongly
corridors.
southern
withponddensities.
in the westernstatesor westernMexico associated
ornorth- Blue-wingedteal densitieswere posiAlberta,
andmoveintoMontana,
ernbreedingareasthrougha westernpor- tivelyrelatedto totalpondsin theeastern
portionof the southernprairieprovinces,
tal (Bellrose1980).
Thisspeciesexhibitedmodestbutcon- northernMontana,all of NorthDakota,
betweenitsdensityand andwesternSouthDakota(Fig.8C).Negsistentcorrelations
were detectedin the
localponddensityexceptin the eastern ative correlations
stratawithinitsrangeandin the
of theareain which northern
portions
andnorthern
these
pondsweresurveyed(Fig.7B;Table3). southernAlbertaprairie,suggesting
of drought-distended to be high in the strataasultimatelocations
Correlations
southwesternstrata(especiallyeastern placedbirds.Blue-wingedteal numbers
Montanaand westernSouthDakota),al- weregreaterin stratain Alaskawhenthe
there. totalnumberof pondswas low (Hansen
thoughthe specieswasuncommon
Green-wingedteal populationsdis- 1960).Stoudt(1971)suggesteddrought
withtotalponds displacement
alsoto the south.
associations
playedpositive
muchof the southernprairie Breedingblue-wingedteal are not
throughout
(Table4).Infact,theyareconprovinces,in northernMontana,and in philopatric
Bellrose(1980)reeasternNorthDakota(Fig.7C).Negative sideredstrongpioneers;
in whichtealnestoccurredin many northern portedseveralinstances
correlations
range
breeding
edwelloutsidetheirnormal
strata.
on homingtendenciesfor in responseto newlyfloodedhabitat.
Information
tealislacking(Table4).We
green-winged
Althoughnot conclusive,the fact that
of birdsarriving correlations
withpondcountsweremore
suggestthatthe majority
in springfromthe southsettlein fairly relatedto averagedensityof blue-winged
goodnumbersin southernstrata,if pond tealthanto latitudesuggeststhatthebirds
therearefavorable.Otherwise donotsimplyfilltheirbreedinghabitatas
conditions
it, butoccupytheprimary
they proceedfarthernorth.Manygreen- theyencounter
BLUE-WINGEDTEAL
B
hi ghest quart i l e
E
th i rd quart i l e
second quartile
:1
lowest quartile
mmeelan
,
r
Emmedian
ZX
r
media
X
,
-
Fig. 8. (A) Average species density by stratum, 1955-81;
density and total pond counts, 1955-81.
1955-81(B) correlation between species density and local pond counts,
NORTHERN SHOVELER
7
j
2
t
thi rd c
_
quar t 1le
-
loifest
I
/
\
-;
hi ghest quart i le E
ticg sec ^
/)
L
J
'
g
/}f:tR't
,
X
1g,90
Re
k
9-
C
_>
B
e
>>
*
Mdmedian
r
/median
E_
r
§
median
-median
r
medlan
r
median
local pond counts, 1955-81;
Fig. 9. (A) Average species density by stratum,1955-81; (B) correlation between species density and
1955-81.
density and total pond counts,
24
WILDLIFEMONOGRAPHS
areafirst.If conditionsthereare not fa- GulfCoast(Bellrose
1980).It appearsthat
usethe
vorable,theymovenorthward,
orpossibly onlyabout40%of springmigrants
southernportalto the breedinggrounds;
backsouth.
Northern Shoveler.-Shovelers reached the remainder
use the westernportalor
highestdensitiesmostlyin theprairieand proceeddirectlyto far-northern
breeding
parklandstrataof the southernprairie grounds.
provinces
andcentralNorthDakota(Fig. Densitiesof thisspeciescovariedwith
9A).Averagedensitiestendedto decline localponddensitiesin the surveyedarea
awayfromthe primarybreedingarea. morecloselythananyotherspecies(Table
Shovelerswintermostlyin California, 3). Correlations
wereconsistentthroughthe west coastof Mexico,and the Gulf outthatarea(Fig.10B),buttendedto be
Coastof Texasand Louisiana(Bellrose largerwherepintailsweremostcommon.
1980).Nearly60%of themprobablyre- Correlations
werehighalsoin westernand
turnin springthroughthesouthern
portal, centralSouth Dakota,althoughpintail
and mostof the rest enterthroughthe densitiestherewerelow.
westernportal.
The pintailexhibitedverystrongposiwithtotalpondsthroughDensities
of thisspeciesvariedmarkedly tivecorrelations
in responseto localpondnumbers(Table out the southernpartof its range(Fig.
3). Correlations
weregreatestwhereshov- 10C).In the northernstrata,associations
elerswerecommon,exceptin strataalong wereclearlynegative.
thenortheastern
edgeof theprimaryarea, Pintailsmayhomeat a highratewhen
arestable(Sowls1955,
whereshovelers
werecommonbutdidnot habitatconditions
Bellrose1980).Henny (1973)suggested
correlatehighlywithponds(Fig.9B).
Therewasa cleardistinctionbetween thatpintailsthatoverflyto the northin 1
the southernpartof the shoveler's
range, yearare likelyto returntherethe next,
whereit tendedtocorrelate
positively
with even if wetlandconditionsin the south
total ponds,and the northernportion, improve.Nonetheless,pintailsgenerally
wherethereverseheld(Fig.9C).Drought- prefertemporarywetlands(Smith1970,
displacedbirdsseemto settlenorthwest, Stewartand Kantrud1973);therefore,
north,andnortheast
of the primaryarea. hominglikelyoccursat a lowrate.Infact,
Although
shovelers
homeata fairlyhigh pintailshaveresponded
opportunistically
rate(Table4), theyrespondpositivelyto to newlyavailablewetlands(Stoudt1971,
1972) and
pondconditions,
particularly
in thesouth- Hochbaumand Bossenmaier
ernandwesternpartof thesurveyedarea. appearto be highlysensitiveto drought
The patternof correlations
is consistent conditions(Crissey1969, Smith 1970,
withthe idea thatthereare 2 pathways Henny1973,DerksenandEldridge1980).
hypothesis
is thatpintailsenintothebreeding
range,onefromthesouth, A reasonable
corridor
respond
the otherfrom the west. Birdsarriving teringalongthesouthern
from eitherdirectionmay settle in the initiallytowetlandconditions
inSouthDain NorthDakota,
southernor westernstrata,if pondsthere kota,nextto conditions
Birds
areadequate;
otherwise,
theymoveon to andeventuallyproceednorthward.
arriving
fromthesouthwest
areinfluenced
the northwest,
north,or northeast.
in southNorthern Pintail. Pintailsreached initiallyby wetlandconditions
very high densitiesin severalnorthern easternAlbertaand southwesternSasIf conditions
therearenotsuitstrataas well as in the southwestern
part katchewan.
of theprairieprovinces
(Fig.lOA).Within able,theymovenorthward.
Manypintails
the prairieandparkland
region,densities migratedirectlytothefarnorth,bypassing
tendedto declineawayfromthe primary the prairiesandparklands
entirely.
Redhead.-This divingduckhadhighbreedingarea.
The pintailis the mostnumerouswin- estdensitiesin theparkland
of theprairie
tering speciesin California.Additional provinces,the mixedforestof southern
pintailswinterin Mexicoand alongthe Saskatchewan,
and the centralDakotas
NORTHERN PINTAIL
r
r
S
a.
m
s
: T
s
/
X
(
T
I
J
t
5
,
fi
{
.
_k
1
.
^
ti
W
->
,
t^l
second quarti le
*
e
.
f
/
E
n
--t2/
^s
W;
__,
X
g
I
;median
I
L
r
__
__
w/median
\
/
r
;Ui
lowest qux
Fig. 10. (A)Averagespecies density by stratum, 1955-81; (e3)correlation between species density and local pond counts, 1955-81;
density and total pond counts, 1955-81.
REDHEAD
median
E251 >0 60
_
w
S
second quartile
2
r
median
r
EJ >0 20 co.40
_
.
,am
L<J
.
lowest quartfle
Fig. 11. (A) Average species density by stratum, 1955-81;
density and total pond counts, 1955-81.
(B) correlation between species density and local pond counts, 1955-81;
OF DUCKS-Johnson
BREEDINGDISTRIBUTIONS
and Grier
27
tendedtode- mainderare distributedin southernstates
(Fig.11A).Averagedensities
cline steadilywith distancefrom these (Bellrose1980).Roughly75%of returning
enterthe breedinggroundvia
canvasbacks
areas.
winteralongthecoastof the southernportalor a southeasternone
Mostredheads
theGulfof Mexico(Bellrose1980).About in the easternDakotas;about25%use the
90%of themreturnto themajorbreeding westerncorridoror fly directlyto areasin
the north.
groundalongsoutherncorridors.
betweendensitiesof this Correlationswith local pond densities
Correlations
speciesand localpondsweremostlylow had a medianof 0.192 (Table3) and were
(Table3), includingstratawith highest consistentacross much of the surveyed
area,includingsome stratain which canredheaddensities(Fig.11B).
Redheadsrespondedpositivelyto total vasbackswere not common(Fig. 12B).In
pondsin onlya few stratain thesouthern Manitoba,correlationswere low, despite
prairieprovincesand in south-centralhigh densitiesof the species. In general,
NorthDakota(Fig. 11C).Negativecor- therewas no relationbetweencorrelation
forseveralstrata coefficientsand averagecanvasbackdenwerecalculated
relations
in the northernpartof its rangeand in sities.
densitieswerepositivelyasIn spiteof lowcor- Canvasback
Montana.
east-central
relationswith localwetlandcounts,red- sociatedwithtotalpondsmostlyin central
headswere less commonin thesestrata Alberta, southern Saskatchewan, and
southwesternManitoba(Fig. 12C). Corwhentotalpondnumberswerehigh.
Redheadsappearto be stronglyphilo- relationswere negativein severalstrataof
patric(Table4), whichmayalsocontrib- the northand in centralCanada.
of therelationbetween Canvasbacksare strongly philopatric
utetotheweakness
aplocal pond countsand numbersof red- (Table4). To someextent,canvasbacks
heads.Hochbaum(1946)suggestedthat pearto occupyhabitatas it is encountered,
the redheadasa divingduck,wasa poor althoughstronghomingtendsto maintain
pioneer.McKnight(1974)disagreed,cit- populationsin the primaryportionof their
ing a rapidbuildupof redheadsin newly breedingrange.Displacementis primarily
to stratain the northand northwest(Fig.
in Utah.
marshland
impounded
Redheaddensitiesdidnotrelateclosely 12C). Dry conditionsin the prairieand
tolocalpondcounts,possiblybecausethey parklandtended to increasecanvasback
prefer the more stable semipermanentnumbersin the northernandnorthwestern
wetlands,whichlluctuatelessin numbers strata.Hansen(1960) also reportedoverthan do temporaryand seasonalponds. flightby this species.
with Lesser Scaup. The distribution of
of associations
thestrength
Moreover,
didnotdisplaya latitudinal scaup was very differentfrom the other
pondnumbers
gradient(Fig.11B).Thus,someredheads species,with highestdensitiesattainedin
seem to respondto pondconditionsen- the northand in 2 stratamostlyin Maniflight,but toba (Fig. 13A). Scaup were uncommon
counteredon theirnorthward
mostmigrateto theirprimaryrange.If southof Canada.Althoughaerialobservers
therearenotsuitable, cannotdistinguishlesserscaupfromgreatwetlandconditions
(seeHansen1960) er scaup (Aythyamarila),it appearsthat
northward
theycontinue
greaterscauppredominatein coastalstrata
or retreatintoMontana.
wasmost of Alaskaand the NorthwestTerritories
canvasback
Canvasback.-The
commonin the parklandregionand in (strata1, 8, 9, 10, 11, and 13), whereas
severalnorthernstrata(Fig.12A).Densi- most of the scaup (an estimated82%)in
tiesdeclinedwithdistanceawayfromthe the interiorstrataarelesserscaup(Bellrose
1980, Boyd 1983). Nonbreeding lesser
primarybreedingarea.
winterin scaupmay summerfar southof theircusAbouthalfof thecanvasbacks
easternstatesand abouta fourthof the tomary breedingareas (Wetmore1920);
populationwintersin the west. The re- therefore,the distributionsuggestedby the
CANVASBACK
<
r7wA
B
highest
quartile
L23 third quartile
r
L _] lowest quartile
|
m
second quartile
i
T
j
I
_
l'_L
/
7
a-
'';
t
8
ii
,
ta
;D
|
8
}'
-
'
{sn median r
t22: >0.60
rza median r
223 >0.20,c0.40
re=n median r
tELj >0.40,<0.60
X w lmedian
L__l <0.20
r
riSr me
ti29 >0
Xm
iiE3 >0
t
pond counts, 1955-81; a
species density and local
(B) correlation between
1955-81;
stratum,
by
density
Fig. 12. (A) Average species
1955-81
density and total pond counts,
-SCAUP
2
hi ghest
X
second
quart i le
quart i le
z
thi rd quart i le
g
median
u
towest
1
>0.40,
quart i le
r
<0 .60
3
>0 .20, <0 .40
g
Lg
<0. 20
[g3
Fig. 13. (A) Average species density by stratum, 1955-81; (B) correlation between species density and local pond countsl 1955-81;
density and total pond counts, 1955-81.
30
WILDLIFEMONOGRAPHS
aerialsurveysmaynotbe the truebreed- nearlyallgadwalls,
blue-winged
teal,redingdistribution.
Also,lesserscaupmigrate heads, and scaup enter their primary
latein springandmaybecountedin strata breeding
rangesfromthesouth.Thesecorsouthof theirultimatedestination
(A.D. ridorsare takenby about75%of greenAfton,pers.commun.).
wingedteal, mallards,and canvasbacks,
Lesserscaupwintermostlyalongthe and by about60%of Americanwigeons
Atlanticand Gulfcoasts(Bellrose1980). and northernshovelers.Onlyabout40%
The fractionof them returningthrough of returningnorthernpintailsuse the
thesouthern
portaltothebreeding
grounds southern
corridor.
probably
exceeds90%.Greater
scaupwinOverall,speciesdensitiestendedto corter alongthe coastsof bothoceans,and relatepositively
withponddensitiesin the
sparinglyin the Gulfof Mexicoand the sametransect.Thisrelationwasstrongest
GreatLakes(Bellrose1980).Theirpri- amongpintails,but also pronouncedin
marymigration
pathways
areeitheralong mallards,
blue-winged
teal,andshovelers.
the PacificCoastor northwestward
across Threespecies,American
wigeon,norththe GreatLakes.
ernshoveler,andnorthern
pintail,tended
Correlations
betweenscaupdensities
and to havehighercorrelations
withpondsin
countsof localpondswerelow (Table3). thosestratawhereeachspecieswasmost
The highestvalueswerecalculatedfor 9 abundant.
Thesespeciesalsoweretheones
stratain thesouthern
partof thesurveyed thatmadelessuseof the southernportal
area,noneof whichhadlargenumbers
of to the breedingrange.
scaup(Fig.13B).Thispatternof correla- Pintaildensitiesweremorecloselyretions may reflectthe presenceof non- latedto totalpondsthantheotherspecies,
breedingscaupin thesouthern
partof the with strongpositivecorrelations
in the
area,alludedto above.
southern
prairieprovinces
andinthenorthLesserscauptendedto correlateposi- centralstates,andnegativecorrelations
in
tivelywithtotalpondsin manystratain thenorthern
strata.Theshovelerandmalcentraland southernCanadaand the lardexhibitedsimilar,butslightlyweaker,
UnitedStates(Fig.13C),includingmany relations.Severalspecies gadwall,wistratain whichthe specieswasnot com- geon,blue-winged
teal,redhead,andlessmon. Negativecorrelations
were found er scaup displayednegativecorrelations
onlyin 3 northernstrataand 1 in Mani- in 1 or morestratain thesouthern
partof
toba,stratathatcontainedlargenumbers theirprimaryrange,suggesting
thatthese
of scaupin mostyears(Fig.13A).
stratamaybe the destination
of droughtLesserscaupare stronglyphilopatric displacedbirds.
(Table4) andonlysomewhataffectedby
Homingrateswerehighestforthe redlocalwetlandconditions.
Nonetheless,
the head,canvasback,
andlesserscaup;modstratashowinghighestcorrelations
with eratelyhigh for the mallard,gadwall,
pondcountsaresouthoftheprimary
range American
wigeon,andnorthern
shoveler;
of the species(Fig. 13A),indicatingthat andlowestfortheblue-winged
teal.Little
somebirds,perhapsnonbreeders,
fill the is knownabouthomingbythepintail,and
habitatastheyencounter
it. Displacementno information
is availableforthe greenappears
tobetothenorthwest
andpossibly wingedteal.
to the east.
General Patterns.-These 10 species Relations with Life History
havehighestdensitiesmostlyin theprairie Characteristics
and parklandof southernCanada.Gadwall and blue-wingedteal had distribu- Valuesof the 2 indicesof life history
tionscenteredsomewhat
moretothesouth; characteristics
andtheiraveragearepreAmerican
wigeon,green-winged
teal,pin- sentedin Table5. As predicted,the negtail, and scauphad morenorthernones; ativeof thefirstindexandtheotherindex
and the canvasback
was centerednorth- variedtogether(r = 0.683,P = 0.03).Beward and eastward.Duringmigration, causetheseindicescovary,it is reasonable
BREEDINGDISTRIBUTIONS
OF DUCKS
Johnsonand Grier
31
5. Indices to life history characteristics: average perto combinethem into a singlemeasure Table
manence values of ponds used by species, median correlation
corresponding
to ther- andK-continuum.coefficient with pond densities, and average index.
Averageindicesindicatethatthe pintail,
mallard,blue-wingedteal, and shoveler
PondperMedian
manence correlation
showthemostfeaturesassociated
withunSpecies
indexa withponds Average
predictable
environments
(i.e.,arethemost
pintail
-0.296
0.547
0.126
r-selected).Gadwall,Americanwigeon, Northern
Mallard
-0.310
0.417
0.053
and green-winged
teal are intermediate,Blue-wingedteal -0.325 0.390
0.032
0.024
andthe divingducksaremostassociated Northernshoveler -0.340 0.388
-0.336
0.228 -0.054
withpredictable
environments
(i.e.,areto- Gadwall teal -0.310
0.199 -0.056
wardthe K-selectedend of the contin- Green-winged
Americanwigeon -0.334
0.143 -0.100
uum).
Canvasback
-0.393
0.192 -0.100
DISCUSSION
Settling Patterns
Redhead
Lesserscaup
-0.384
-0.389
0.154
0.145
-0.115
-0.122
a From
Stewart
andKantrud
(1973);averagepermanence
(cropland
ponds= 2, seasonal
ponds= 3, semipermanent
pondsandreservoirs
=
4) times-0.1.
We foundevidencethatducksexhibit
all 3 of the hypothesized
patternsof settling.Ourliterature
reviewsuggested
that, thewintering
grounds
orduringspringmiwithinspecies,homingwas morepreva- gration.Becausehabitatneedsduringthe
lent amongadult femalesthan among breedingseasonarelesscriticalfor males
males(forthefew speciesin whichmales thanfor females,thereis relativelylittle
havebeenstudied)oryearlings.
Thatadult advantage
to malesin homingto familiar
femalesare morephilopatricthanadult sites.Instead,themalefollowsthehoming
malesis consistentwiththe seasonalmo- female(McKinney
1965).Greaterphilonogamyof thesespeciesand the greater patryby femalesalsois generallyconsisroleofthefemaleinparental
care(Rohwer tent with the predictionof Greenwood's
and Anderson1988);maledispersalmay (1980)hypothesis
andthe socialstructure
be largelya consequenceof when and of duckpopulations.
Thefemaledoesnot
wheretheducksformpairbonds.Because selecta mateon thebasisof thequalityof
of differences
in parentalcare,the repro- his territory.Accordingly,Greenwood's
ductivesuccessof malesis limitedby ac- (1980)hypothesis
predictsthatfemalesare
cess to fertilizablefemales,whereasthe morelikelyto homethanmales.
reproductive
successof femalesis limited Forthe mallard,gadwall,andnorthern
byenvironmental
resources
neededforre- shoveler,at least,thereis evidencethat
production.Becausefemalesspend far hensthataresuccessful
inhatching
a clutch
moretime thanmalesat nestsand with are moreproneto homethanthosethat
young,they are exposedto greaterpre- arenot (DotyandLee 1974;Mihelsons
et
dationrisk(Sargeant
1972,Sargeant
et al. al. 1986;J.T. Lokemoen,
H. F. Duebbert,
1984),whichsubsequently
resultsin a pre- andD. E. Sharp,Reproductive
strategies
ponderanceof malesin the population of prairiemallards,gadwalls,and blue(Johnson
andSargeant
1977).Thus,allfe- wingedteal,unpubl.manuscript,
U.S.Fish
malescan generallyfinda mate,andse- andWildl.Serv.,Jamestown,
N.D.).This
lectionshouldfavorthosethatarebestable phenomenon
couldbe viewedas a "playto findqualitybreedinghabitat,whichin- the-winner"
approach
to settlingin a habcludessecurenestsitesandaccessto food itat;thereis noadvantage
to be gainedby
duringegglayingandincubation.
Females changinga habitatthathasbeensuccessmaytendtohomebecausefamiliarity
with ful. In fact,somestudieshaveshownina breedingareashouldfacilitateaccessto creasedreproductive
successamongbirds
bothfoodanda securenestsite.Formales, that usednest sitesin whichthey were
not all of whichcan mate,the selective previouslysuccessful(Dow and Fredga
pressure
to obtaina femaleconfersanad- 198S,BlancherandRobertson
1985).
vantageto pairingas earlyas possible,on
Contrasting
the10species,homingseems
2
WILDLIFEMONOGRAPHS
mostprevalentamongredhead,canvas- to winteringgrounds,whichis consistent
back, lesser scaup, mallard,gadwall, with predictionsof Fretwelland Lucas
Americanwigeon,and shoveler.These (1969).Nuddscitedunpublished
workof
speciesgenerallyusemorestablewetlands H. A. Kantrud,who foundthat correlathanthe others(Smith1971,Stoudt1971, tionsbetweenponddensityanddensityof
Stewartand Kantrud1973);theiruse of breeding-duck
pairsdecreasedwith inthesemorestablehabitatswouldpredict creasing
latitude.
Thatresult,however,
was
a greaterhomingtendency(e.g., Wiens basedon countsof all speciescombined.
1976).
Incontrast,
the10speciesexamined
inthis
All speciesexhibitedopportunistic
set- reportfilledthe habitatsomewhatdiffertlingto someextent,in thattheirnumbers ently. Four of them gadwall,greenfluctuatedwith pond numbers.Such winged teal, canvasback,and lesser
movementis especiallyimportant
in spa- scaup diddisplaya latitudinal
declinein
tiallyheterogeneous
habitats(Gauthreauxcorrelation
withpondnumbers,although
1982),such as wetlandsin the prairie, noneof thesespecieswascloselyassociated
wheredroughtis oftena localphenome- withpondcounts.The3 speciesthatwinnon. Mostnotableas opportunists
were terinlargenumbers
inthewesternstatespintail,blue-wingedteal, mallard,and American
wigeon,northern
shoveler,and
shoveler.Opportunism
is favoredwhen northernpintail-showedhigh correlahabitatconditions
areunpredictable
from tionswithpondnumbersin southwestern
1 yearto the next.The speciesusingthe strata,andthelatter2 didin severalsouthleastpredictablewetlandhabitatsis the ern strataas well.Althoughit is possible
pintail,which also is the speciesmost thatmanybirdsactuallyhomebut then
stronglycorrelated
withpondcounts.Div- backtrack
to areasencountered
earlierin
ingducks,at theotherextreme,usemore migration,it seemsclearthat largesegpredictablewetlands;correspondingly,mentsof thepopulations
of these7 species
thesespecieswerelesscloselytiedto pond respondto pondconditionsas they encounts.Moreconclusiveevidencefor op- counterthem.
portunism
is a patternof strongestcorre- Theremainingspecies mallard,bluelationswithpondcountsin stratafirsten- wingedteal,andredhead appearto recountered
byreturning
birdsinthespring. spondmoredirectlyto wetlandconditions
Thispatternwasapparentfor American in the primaryportionof theirbreeding
wigeon,green-wingedteal, and canvas- ranges,as is evidentfromhighestcorreback and was pronouncedfor gadwall, lationswith pondnumbersthere.These
shoveler,andpintail.
findings
fittheideal-free
distribution
modFlexiblesettling,indicatedbyoverflight el of FretwellandLucas(1969),whoprein dry years,wasshownby mostspecies dictedthatthe"mostsuitable"
habitatfills
whenappropriate.
Droughtdisplacementfirstandthatsomeindividuals
areforced
wasgenerallyto the northor northwest, tolesssuitablehabitats.
Forseveralspecies
as suggested earlier by Hansen and of ducks,the mostsuitablehabitatmay
McKnight(1964), Crissey(1969), and not be thatclosestto winteringareas,as
Smith(1970).Someoverflight
tothenorth- Nudds(1978)conjectured,
butratherthat
east was indicatedfor mallard,gadwall, habitatintheprimary
partof thebreeding
blue-winged
teal,green-winged
teal,red- range.
head,andcanvasback.
ForAmericanwigeon,manyof whichenterfromthewest, Relations to Life History
droughtdisplacementto the east and Characteristics
southeast
wasseen.
Wefoundthat2 independent
indicesto
suites
of
life
history
characteristics
(these
Fillingthe Habitat
canbe equallyconsidered
as indicesof rNudds(1978,1983)suggestedthatthe andK-selection)
variedtogether.Average
habitatof ducksbecomesfilledfirstclosest valuesindicated
thatpintail,mallard,
blue-
BREEDINGDISTRIBUTIONS
OF DUCKS
Johnsonand Grier
33
wingedteal, and shovelerhavethe most amongduckspeciesis bestmadein terms
featuresassociatedwith unstableor un- of annualsurvivalrates,whichcanbe espredictable environments. Gadwall, timatedfromlong-termbandingstudies.
American
wigeon,andgreen-winged
teal D. H. Johnson,
J. D. Nichols,and M. D.
wereintermediate,
and attributesof the Schwartz(Breedingdynamicsof ducks,
divingduckswereassociated
withthe use unpubl.manuscript
presented
atEcol.and
of stableor predictable
environments. Manage.of BreedingWaterfowlSymp.,
The conclusions
of 3 otherstudiesare Winnipeg,Manit.,Canada,18-22 Aug
consistentwith ours.For one, Patterson 1987)reviewedstudiesthatusedmodern
(1979)suggested
thathighcorrelation
coef- statisticalmethodsof estimatingsurvival
ficientsbetweenducknumbersandpond rate.They foundsomevariationamong
numbers
intheCanadian
prairieandpark- species,but not all of the 10 specieshad
landsindicatedan opportunistic
response beenstudied,andestimatesavailableperto habitatconditions
associated
withr-se- tainedto differentgeographic
areas,time
lection.He also determinedcorrelationsperiods,
andbanding
seasons,
makingthem
betweenchangesin ducknumbers
andthe difficultto compare.
Moreover,
allstudied
previouspopulation
count,largepositive populations
hadbeensubjectedto recrecorrelations
interpretedas evidencefor ationalhunting,so thatestimatesmayno
competitive
self-regulation
associated
with longerreRectsurvivalratesundercondiK-selection.He concludedthat, among tionsto whichthe specieshadadapted.
these10 duckspecies,the mallard,blueComparisonsof reproductiveeffort
wingedteal, pintail,and shovelerwere amongspeciesareoftenmadeonthebasis
r-selecteddabblers.Americanwigeon, of ageat firstreproduction
andclutch(or
gadwall,andgreen-winged
teal wereK- litter)size.Forwaterfowl,
totalnumberof
selecteddabblers,and redhead,canvas- eggslaidduringa breedingseasonis more
back,andlesserscaupwerefurtheralong relevantthanclutchsizebecauseit takes
the spectrumtowardK-selection.
differingrenestingprobabilities
into acBailey(1981)comparedmallardand count.
canvasback
life historieswith datagath- We reviewedPalmer(1976)and BelleredinSaskatchewan
byStoudt(1971).He rose(1980)forinformation
on ageat first
concludedthatthe mallardwasmoreof reproduction.
The4 specieswe indicated
anr-selectedspeciesthanthecanvasback,weremostr-selected
(pintail,
mallard,
bluein thatthe mallardrespondedmoreop- wingedteal,andshoveler)
typicallybreed
portunistically
to improvinghabitatand asyearlings,
asdoesAmerican
wigeon,an
acocepteda greaterriskof reproductiveintermediate
species.Of theother2 interfailure.
mediatespecies,the gadwalloftendefers
Finally,Vickery
andNudds(1984)tested breedinguntilits secondyear,andno infor density-dependent
relationsin duck formationwas presentedon the greenpopulations
on studyareasin Albertaand wingedteal.Largenumbers
of the3 most
Saskatchewan.
Theyfoundthatgadwall, K-selected
species canvasback,
redhead,
northernshoveler,redhead,canvasback,andlesseerscaup donotbreeduntilthey
andlesserscaupshowedsomedensityde- are 2 yearsold. Thus,age at firstrepropendence,characteristic
of K-selection. ductionparallels
the otherattributes
of rTheremaining
5 speciesdid not.
andK-selection
thatwe examined.
It wouldbeworthwhile
alsotoarraythe
Totaleggslaidin a breedingseasonis
10 speciesaccordingto otherassociates
of moredifficultto addressbecauseit inr- andK-selection,
notablylongevityand volvesclutchsizeandthenumberof nestreproductiveeffort. Speciesconsidered ing attemptsmade.Further,clutchsize
morer-selectedtendto haveshorterlife- typicallydeclineswith later nestingatspansbut higher fecundity;K-selected tempts,and bothclutchsize and probaspeciestypicallylive longerbut produce bilityof makinga nestingattemptdepend
feweryoung(e.g.,Pianka1974).
on theconditionof thebreedingbirdand
A comparisonof ecologicallongevity its habitat(D. H. Johnson,
J. D. Nichols,
4
WILDLIFEMONOGRAPHS
suchas
andM. D. Schwartz,Breedingdynamics Evenwithoutnewinformation,
of ducks,unpubl.manuscript
presented
at satelliteimageryprovidinghabitatconEcol.andManage.
of BreedingWaterfowl ditions,an understanding
of the factors
canimSymp.,Winnipeg,Manit.,Canada,18-22 thatdetermineduckdistributions
Aug1987).Keith(1961)estimatedfor an proveestimates
of ducknumbers.
Johnson
Albertastudyareaboth clutchsize and (1981,1986)showedhowauxiliaryinfornumbers
of nestsperfemale.Weusedthe mationcouldbe employedto developesproductof those2 valuesasanestimateof timators,knownas empiricalBayesestitotaleggslaid(exceptforthecanvasback,mators,thataveraged20%moreaccurate
whichhadlimitedsamplesize,andpintail thancustomaryestimators.
Auxiliaryinand American wigeon, which Keith formation
took1 of thefollowing3 forms:
thoughtwerenotadequately
represented).countsof a givenspeciesin a particular
For the 7 remainingspecies,total esti- stratumduringpreviousyears,the pond
matedeggproduction
wascorrelated
with countin thatstratumduringthe yearof
ourindexof r- andK-selection
(r= 0.705, thesurvey,andthetotalnumberof ponds
P = 0.077;Table5).
withwaterwithinthepond-surveyed
area.
Theanalyseswe present,the otherevi- Although
theprevious
countwasgenerally
dencewe review,andthe reportsof Pat- aneffectiveformof auxiliary
information,
terson(1979),Bailey(1981),andVickery estimatesfor somespeciesin somestrata
andNudds(1984)donotjustifyanycausal werebetterwitheitherthe localor conimplications,but suggestan association tinentalpond count. These differences
amongcertainlife historyfeatures,in- amongspeciesandstratacorresponded
to
cludingthepredictability
of favoredhab- findingsof thisreport.
itats.
Oneof themainpurposes
of estimating
breedingpopulations
of ducksis topredict
the fall flightso thathuntingregulations
Management Applications
can be established
consistentwithanticiEffective managementof waterfowl patednumbers
of available
ducks.Forany
populations
is enhancedby a basicunder- species,the fall flightdependson the size
standingof the distributions
of the birds of the breedingpopulationand its proandthe factorsthataffectthosedistribu- ductivity.Productivity
in turndependson
tions.Ona continental
basis,forexample, severalfactorsand is variablefromarea
mucheffortis expendedin attemptsto to areaandfromyearto year.Again,an
estimatethe size of eachyear'sbreeding understanding
of thedistribution
of breedpopulation
of key speciesof ducks.Wa- ing ducksis important
in predicting
their
terfowlsurveysto obtainsuchestimates productivity.
For example,greaterproarelabor-intensive
andcostly.Conversely, ductivitymay be expectedif mostmalextensivehabitat informationmay be lardsaredispersed
acrossbreeding
habitats
gatheredinexpensively
with satelliteim- in theprairiesandparklands,
andif pond
agery,as a by-product
of otheractivities. conditions
therearegood,thanif theyare
By knowinghow ducksdistributethem- displacedto morenorthern
strata.Results
selvesin response
to habitatconditions,
es- in this reportcan be usedto refineprepeciallythoseof wetlands,surveyscould dictionsbeyondsuchobviousgeneralities.
be designedin a moreoptimalfashion. Ona long-termbasis,thereis muchinSampling
intensitycouldvaryannuallyto terestin knowingwhetherpopulations
of
placemoreeffortin stratawherethehab- ducksare increasing,fluctuatingabout
itat information
predictsthe presenceof somelong-termmean,or declining.Demore ducks,for example.Resultspre- termining
trendsis notas straightforward
sentedin thisreportwillbe usefulin plan- as it wouldseembecauseof the variable
ningextensivewaterfowl
surveys,
butthey distributions
of the speciesand the fact
alsowarnthatthe 10 speciesvary,so that thatnumbers
of ducksarerelatedto numan optimaldesignmustbe a compromise bersof pondswith water.Analyseslike
amongthe speciesof interest.
thoseincludedherecan help clarifythe
OF DUCKSBREEDINGDISTRIBUTIONS
Johnsonand Grier
35
male strategiesof reproduction.Ph.D. Thesis,
situation.A studyof mallardtrendswas
281pp.
Univ.Minnesota,Minneapolis.
by JohnsonandShaffer(1987). ANONYMOUS.
presented
operatingprocedures
1977. Standard
such
morelocalsituations,
Considering
foraerialwaterfowlbreedinggroundpopulation
as a wetlandmanagermightdealwith,it
andhabitatsurveys.U.S.Fishand Wildl.Serv.,
Off.of Migr.Bird Manage.81pp.
how
is equallyimportantto understand
approachtoprobR.O. 1981. A theoretical
ducksdistributethemselves.This infor- BAILEY,
in waterfowlmanagement.Trans.North
lems
mationwill help assessthe ef:fectsof onAm.Wildl.and Nat. Resour.Conf.46:58-71.
the-groundmanagementpractices.For BELLROSE, F. C. 1980. Ducks,geeseandswansof
NorthAmerica,Thirded. StackpoleBooks,Harexample,it maybe feasibleto attemptto
risburg,Pa. 540pp.
developa largebreedingpopulationof
1970. Migrational
, AND R. D. CROMPTON.
divingducksor someotherspecieswith
behaviorof mallardsand blackducksas deterhigh ratesof homingby attractingpairs
minedfrombanding.Ill. Nat. Hist.Surv.Bull.
Surviving 30:167-234.
and ensuringhigh production.
adultfemalesand manyof theirfemale BISHOP,R.A., D. D. HUMBURG,ANDR. D. ANDREWS.
1978. Survivalandhomingof femalemallards.
offspringwouldbe likelyto returnto the
Wildl.Manage.42:192-196.
J.
area,if the breedinghabitatwerestable BLANCHER,
1985. Site
P. J., ANDR. J. ROBERTSON.
fromyearto year.Thiseffortwouldbe
consistencyin kingbirdbreedingperformance:
implicationsfor site fidelity.J. Anim.Ecol.54:
less likelyto succeedif directedtoward
1017-1027.
teal,howpintailsorblue-winged
northern
homingof male
R. J. 1978. Migrational
ever,becauseof theirreducedhomingten- BLOHM,
gadwallsto breedinggrounds.Auk95:763-766.
dencies.The developmentof attractive
1979. Thebreedingecologyof thegadwall
partsof thebreeding in southernManitoba.Ph.D.Thesis,Univ.Wiswetlandsin southern
consin,Madison.177pp.
range would more likely attractthese
of r- andK-selection
1984. Restitution
naturalselecof density-dependent
model
a
as
Beyondtheireffortsto improveduck
tion.Annu.Rev.Ecol.Syst.15:427-447.
canfill BOYD,H. 1983. Intensiveregulationof duckhuntmanagers
andpopulatiorls,
habitats
ing in NorthAmerica:its purposeand achievea vitalscientificrole.Mostof theirmanments.Can.Wildl.Serv.Occas.Pap. 50. 24pp.
agementactivitiescanbe viewed,notonly
B. K., AND D. A. BOAG. 1977. Reproasa positiveactionfortheducks,butasa CALVERLEY,
ductivepotentialin parklandandarctic-nesting
scientifictest.Theoriessuchas presented populationsof mallardsand pintails(Anatidae).
to
of responses
heregeneratepredictions
Can.J. Zool. 55:1242-1251.
specificchanges.Theycan be appliedto COULTER,M. W., AND W. R. MILLER. 1968. Nestingbiologyof blackducksandmallardsin northactionsandresultsusednot
themanager's
ern
NewEngland.VermontFishandGameDep.
the duck
species.
onlyto evaluatethe impacton
but alsoto evaluatethe scipopulations,
entifictheory.We hope that this report
notonlyprovidessomeguidanceformanactionsthatwill test
agers,butstimulates
its predictions.
BOYCE,M. S.
Bull. 68-2. 74pp.
CRISSEY,W. F. 1969.
Prairiepotholesfroma continentalviewpoint.Pages161-171 in Saskatoon
wetlandsseminar.Can.Wildl.Serv.Rep.Ser.6.
DERKSEN, D. V., AND W. D. ELDRIDGE. 1980.
of pintailsto the Arctic
Drought-displacement
coastplain, Alaska.J. Wildl. Manage.44:224229.
DOTY,H. A., ANDF.
CITED
LITERATURE
G. 1979. The populationecologyof
an isolatednestingpopulationof redheads(Aythya americana). Ph.D. Thesis,CornellUniv.,
Ithaca,N.Y.211pp.
ANDERSON,D. R., AND C. J. HENNY. 1972. Populationecology of the mallard:I. A review of
andmigrapreviousstudiesandthe distribution
tion frombreedingareas.U.S. Fish and Wildl.
Serv.Resour.Publ.105. 166pp.
of breeding
ANDERSON,M. G. 1985. Socialbehavior
(Aythya valisineria): male and fecanvasbacks
ALLISTON,W.
B. LEE. 1974. Homingto nest
basketsby wildfemalemallards.J. Wildl.Manage.38:714-719.
andnatal
DOW, H., ANDS. FREDGA. 1983. Breeding
dispersalof the goldeneye,Bucephala clangula.
J. Anim.Ecol.52:681-695.
1979.
D. M.JANKE.
DWYER,T. J.,G. L. KRAPU,AND
Use of prairiepotholehabitatby breedingmallards.J.Wildl.Manage.43:526-531.
on carryingcapacity
DZUBIN,A. 1969. Comments
of smallpondsfor ducksand possibleeffectsof
densityon mallardproduction.Pages138-160
wetlandsseminar.Can.Wildl.Serv.
in Saskatoon
Rep.Ser.6.
FRETWELL,S. D., AND H. L. LUCAS,JR. 1969. On
WILDLIFEMONOGRAPHS
36
, AND T. L. SHAFFER. 1987. Are mallards
territorial
behaviorandotherfactorsinfluencing
habitatdistributionin birds:I. Theoreticaldedeclining in North America?Wildl. Soc. Bull. 15:
velopment.ActaBiotheoret.19:17-36.
340-345.
GATES,J. M. 1962. Breedingbiologyof the gadwall JOHNSON,D. J. 1978. Age-relatedbreedingbiology
in northernUtah.WilsonBull.74:43-67.
of the redhead duck in southwesternManitoba.
GAUTHREAUX,S. A., JR. 1980. The influencesof
M.S. Thesis, Texas A&M Univ., College Station.
long-termandshort-term
climaticchangesonthe
47pp.
dispersalandmigrationof organisms.
Pages103- KEITH, L. B. 1961. A study of waterfowl ecology
174in S. A. Gauthreaux,
Jr.,ed. Animalmigraon small impoundmentsin southeasternAlberta.
tion,orientation,
andnavigation.
AcademicPress,
Wildl. Monogr.6. 88pp.
New York,N.Y.
KRAPU,G. L. 1979. Nutrition of female dabbling
1982. The ecologyand evolutionof avian
ducksduring reproduction.Pages 59-70 in T. A.
migrationsystems.Pages93-168in D. S.Farner,
Bookhout, ed. Proc. 1977 Symp. Midwest Fish
J. R. King,andK. C. Parkes,eds.Avianbiology.
and Wildl. Conf., Madison, Wis. North Cent.
Vol.6. AcademicPress,New York,N.Y.
Sect., The Wildl. Soc.
GREENWOOD,P. J. 1980. Matingsystems,philo- LEITCH,
W. G., ANDR. M. KAMINSKI.
1985. Longpatryanddispersal
in birdsandmammals.Anim.
term wetland-waterfowltrends in Saskatchewan
Behav.28:114s1162.
grassland.J. Wildl. Manage. 49:212-222.
, AND P. H. HARVEY. 1982. The nataland MAcARTHuR,
R. H., ANDE. O. WILSON.1967. The
breedingdispersal
of birds.Annu.Rev.Ecol.Syst.
theory of island biogeography. Princeton Univ.
13: 1-21.
HANSEN, H. A.
Press, Princeton, N.J. 203pp.
Changedstatusof several MARTIN,
F. W., R. S. POSPAHALA,
ANDJ.D. NICHOLS.
speciesof waterfowlin Alaska.Condor62:1361979. Assessmentand population management
1960.
of North American migratorybirds. Pages 187239 in J. Cairns,Jr.,G. P. Patil,and W. E. Waters,
eds. Environmental biomonitoring, assessment,
prediction,and management certain case studies and relatedquantitativeissues.Statisticalecology. Vol. 11. International Cooperative Publ.
House, Fairland, Md.
HOCHBAUM,G. S., AND E. F. BOSSENMAIER.1972.
MARTINSON,
R. K., AND C. F. KACZYNSKI.
1967.
Responseof pintailsto improvedbreedinghabFactors influencing waterfowl counts on aerial
itat in southernManitoba.Can. Field-Nat.86:
surveys,1961-66. U.S. Fish and Wildl. Serv.Spec.
79-81.
Sci. Rep. Wildl. 105. 78pp.
HOCHBAUM,H. A. 1946. Recoverypotentialsin
McHENRY,M. G. 1971. Breeding and post-breedNorthAmericanwaterfowl.Trans.NorthAm.
ing movements of blue-winged teal (AnasdisWildl.Conf.11:403-418.
cors) in southwestern Manitoba. Ph.D. Thesis,
HOLM, E. R., AND M. L. SCOTT. 1954. Studieson
Univ. Oklahoma,Norman. 76pp.
the nutritionof wild waterfowl.N.Y. Fish and McKINNEY,
F. 1965. Spacingand chasingin breedGameJ. 1:171-187.
ing ducks. Wildfowl 16:92-106.
IMMELMANN,
K. 1973. Roleof the environment
in McKNIGHT,
D. E. 1974. Dry-land nesting by redreproductionas sourceof "predictive"inforheads and ruddy ducks. J. Wildl. Manage. 38:
mation. Pages 121-147 in D. S. Farner,ed.
112-119.
Breedingbiologyof birds.Nat.Acad.Sci.,Wash- MIHELSONS,H., A. MEDNIS,ANDP. BLUMS. 1986.
ington,D.C.
Populationecology of migratoryducksin Latvia.
JOHNSGARD,
P.A. 1975. Waterfowlof NorthAmerZinatnePubl.House,Riga,Latvia,U.S.S.R.112pp.
ica. IndianaUniv.Press,Bloomington.
575PP.
MURDY,H. W. 1966. When the prairies go dry.
JOHNSON,
D. H. 1980. The comparisonof usage
Naturalist17:8-13.
andavailabilitymeasurements
forevaluatingre- NUDDS,T. D. 1978. Comments on Calverley and
sourcepreference.Ecology61:65-71.
Boag's(1977) hypothesison displaced ducks and
1981. Improved populationestimates
an evolutionaryalternative.Can.J. Zool.56:2239throughthe use of auxiliaryinformation.Pages
2241.
436-440in C. J.RalphandJ. M. Scott,eds.Es1983. Niche dynamics and organizationof
timatingnumbersof terrestrial
birds.Studiesin
waterfowlguilds in variableenvironments.EcolavianbiologyNo. 6. CooperOrnithological
Soc.
ogy 64:319-330.
and AllenPress,Inc.,Lawrence,Ks.
PALMER,R. S. 1976. Handbookof North American
1986. EmpiricalBayesestimatesof breedbirds. Vols. 2 and 3. Yale Univ. Press, New Haing populations
of NorthAmericanducks.Ph.D.
ven, Conn. 521 and 560pp.
Thesis,NorthDakotaStateUniv.,Fargo.187pp. PATTERSON,J. H. 1979. Can ducks be managed by
, AND A. B. SARGEANT.1977. Impactof red
regulation?Experiencesin Canada.Trans.North
foxpredationon thesexratioof prairiemallards.
Am. Wildl. and Nat. Resour.Conf. 44:130-139.
U.S. Fish and Wildl.Serv.Wildl.Res. Rep. 6. PIANKA,E. R. 1974. Evolutionaryecology. Harper
56pp.
& Row, New York, N.Y. 356 pp.
137.
, AND D. E. MCKNIGHT. 1964.
Emigration
of drought-displaced
ducksto the Arctic.Trans.
NorthAm.Wildl.Conf.29:119-127.
HENNY, C. J. 1973. Droughtdisplacedmovement
of NorthAmericanpintailsintoSiberia.J.Wildl.
Manage.37:23-29.
OF DUCKSDISTRIBUTIONS
BREEDING
Johnsonand Grier
37
Stackpole Co.,
SOWLS, L. K. 1955. Prairie ducks.
Inst.,WashManage.
Wildl.
and
Pa.,
Harrisburg,
II.
mallard:
1974. Population ecology of the
193pp.
D.C.
ington,
Breedinghabitatconditions,size of the breeding
1977. The evolution of life history
STEARNS, S. C.
populations, and production indices. U.S. Fish
of the theory and a review of
critique
a
traits:
73pp.
115.
Publ.
and Wildl. Serv. Resour.
Ecol. Syst. 8:145-171.
Rev.
Annu.
data.
the
POSTON,H. J. 1969. Home range and breeding
1958.
EVANS.
STEWART, R. E., A. D. GEIS,AND C. D.
State
Utah
Thesis,
M.S.
shoveler.
the
of
biology
populations and hunting kill of
of
Distribution
Univ., Logan. 86pp.
the canvasback.J. Wildl. Manage. 22:333-370.
III. 1979. PopH. R., ANDT. A. PARKER,
PULLIAM,
1973. Ecologicaldis, AND H. A. KANTRUD.
ulationregulationof sparrows.Fortschr.Zool.25:
populations in
waterfowl
breeding
of
tribution
137-147.
Wildl. Manage. 37:39-50.
J.
Dakota.
North
Fe1988.
ANr)ERsoN.
G.
M.
F. C., AND
ROHWER,
J. H. 1971. Ecological factors affecting
male-biasedphilopatry,monogamy,and the tim- STOUDT,
productionin the Saskatchewanparkwaterfowl
ing of pair formation in migratory waterfowl.
U.S. Fish and Wildl. Serv. Resour. Publ.
lands.
5:187-221.
Ornithol.
Current
99. 58pp.
A. B. 1972. Red fox spatial characterSARGEANT,
D. L. 1971. Populationecology of lesser
isticsin relationto waterfowl predation.J. Wildl. TRAUGER,
(Aythya affinis) in subarctic taiga. Ph.D.
scaup
Manage. 36:225-236.
Iowa State Univ., Ames. 118pp.
Thesis,
1984.
EBERHARDT.
T.
R.
AND
, S. H. ALLEN,
L., ANDT. D. NUDDS. 1984. Detection
W.
VICKERY,
midconin
ducks
Red fox predationon breeding
effects on annualduck cendensity-dependent
of
41pp.
tinent North America. Wildl. Monogr.89.
65:96-104.
Ecology
suses.
INC. 1982. SAS user s guide: statisSASINSTITUTE,
A. 1920. Observationson the habits of
WETMORE,
tics. SAS Inst., Inc., Cary, N.C. 584pp.
Lake Burford,New Mexico.Auk37:221at
birds
Wetlands
SHAW,S. P., ANDC. G. FREDINE.1956.
247.
of the United States. U.S. Fish and Wildl. Serv.
WIENS,J. A. 1976. Populationresponsesto patchy
Circ. 39. 67pp.
environments.Annu. Rev. Ecol. Syst. 7:81-120.
waaffecting
factors
SMITH,A. G. 1971. Ecological
U.S.
terfowl productionin the Albertaparklands.
Received 18 January 1988.
Fish and Wildl. Serv. Resour.Publ. 98. 49pp.
Accepted 17 June 1988.
breeding
pintail
of
Response
SMITH, R. I. 1970.
34:
Manage.
Wildl.
populations to drought. J.
943-946.
ANDC. J. HENNY.
R. S., D. R. ANDERSON,
POSPAHALA,