Download temporal changes in habitat selection and nest spacing in a colony

TEMPORAL
AND
CHANGES
NEST
ROSS'
IN HABITAT
SPACING
AND
SELECTION
IN A COLONY
LESSER
SNOW
OF
GEESE
M. ROBERT MCLANDRESS
Divisionof WildlifeandFisheries
Biology,Universityof California,
Davis, California95616USA
ABSTRACT.--Istudied the nesting colony of Ross' Geese (Chen rossii)and Lesser Snow
Geese (C. caeruIescenscaeruIescens)at Karrak Lake in the central Arctic of Canada in the
summerof 1976.Relatedstudiesindicatedthat this colonyhad grownfrom 18,000birdsin
1966-1968to 54,500 birds in 1976. In 1976, geesenestedon islands that were used in the
late 1960'sand on an islandand mainlandsitesthat were previouslyunoccupied.
Average
nestdensityin 1976wasthree-foldgreaterthan in the late1960's.Consequently,
the average
distanceto nearestneighborsof Ross'Geesein 1976was half the averagedistancedetermined 10 yr earlier.
The mean clutchsize of Ross'Geesewas greaterin islandhabitatswhere nestdensities
were high than in lesspopulatedisland or mainlandhabitats.The averagesize of Snow
Gooseclutchesdid not differ significantlyamongislandhabitatsbut was largerat island
than at mainland sites. Large clutcheswere most likely attributableto older and/orearlier
nesting females. Habitat preferencesapparentlydiffered between species.Small clutches
presumablyindicatedthat younggeesenestedin areaswhere nestdensitieswere low. The
establishment
of mainlandnestingat KarrakLakeprobablybeganwith youngSnowGeese
usingperipheralareasof the colony.YoungRoss'Geesenestedin sparselypopulatedhabitats on islandsto a greaterextent than did Snow Geese. Ross'Geesealso nestedon the
mainlandbut in lower densitiesthan Ross'Geesenestingin similar islandhabitats.Successfulnestswith the largerclutcheshad closerconspecific
neighborsthan did successful
nestswith smallerclutches.
The speciescomposition
of nearestneighbors
changedsignificantlywith distancefromSnowGoosenestsbut not Ross'Goosenests.Nestingsuccess
was
not affectedby the species
of nearestneighbor,however.Because
theyhavecomplementary
antipredatoradaptations,Ross'and Snowgeesemay benefitby nestingtogether.Received
13 May 1982, accepted3 January1983.
MANY Arctic nesting speciesof geese (An-
species (see Cody 1973, Bedard 1976, Burger
serini) are sympatricand sharenumerouslife- 1981 for reviews).
history traits (see Delacour 1954, Johnsgard
The populationof Ross'Geesenumberedless
1978, Bellrose 1980, Owen 1980 for reviews). than 5,000 in the early 1950's,but by 1976 it
Despite these similarities, somespeciesare colonial (see definition in Gochfeld 1980), whereas othersare dispersednesters.Two exclusively colonialspeciesare Ross'Geese(Chenfossil)
and Snow Geese (Chen caerulescens).
The nesting distribution of Lesser Snow Geese (C. c.
caerulescens)
is disjunct but includes colonies
had increased to more than 100,000 birds
(McLandress 1979). In contrast, there are several million Lesser Snow Geese (Owen 1980),
but changesin populationsizeshave been inadequatelymeasured(seediscussionin Kerbes
1975).My objectivewas to study changesin a
nesting colony containing both species,relaacross most of arctic North America and a coltive to the populationexpansionof Ross'Geese,
ony on WrangelIsland in Siberia(Bellrose1980). and to determine interspecificeffectson nestNearly all Ross'Geese, however, are restricted ing distribution and success.
to the centralArctic of Canada,where they nest
in colonieswith Lesser Snow Geese (Ryder
STUDY AREA AND RELATED STUDIES
1969a, 1971a).Multispeciescoloniesare fairly
common among birds (Burger1981); these asThe study was conductedin the summer of 1976
semblageshave been used as a basis for sug- at Karrak Lake (67ø15'N, 100ø15'W;Fig. 1), the site
gesting complexnesting relationshipsthat in- of the largestgoosecolonyin the QueenMaud Gulf
cludeboth competitionand cooperationamong Lowlandsof the centralArctic (Ryder 1969a).Based
335
The Auk 100:335-343.April 1983
336
M. ROBERT
MCLANDRES$
[Auk,Vol. 100
Fig. 1. The KarrakLake nestingcolonyof Ross'and LesserSnow geese(afterRyder 1972).Blackareas
were nestingislandsusedby geese1966-1968and 1976,hatchedareaswere "expansion"sitesoccupiedby
1976, and stippledislands(lower right) were used by nestinggeesein 1966-1968but not in 1976.
on an aerial photographic survey of Karrak Lake,
Kerbes et al. (1983) estimatedthat the colony contained 54,500nestinggeesein 1976.A previousstudy
of the same colonywas conductedfrom 1966to 1968
(Ryder 1972), when colony size was estimatedto be
18,000 birds (visual aerial survey, Ryder 1969a). Although the size of the Karrak Lake colonytripled in
this 10-yr interval,speciescompositionchangedonly
slightly (67% Ross', 33% Snow, Ryder 1969a;60%
Ross', 40% Snow, Kerbes et al. 1983). Geese nest-
ed only on islandsin the mid-1960's,but by 1976the
colony had expanded to include nearby mainland
areasand a previously unoccupiedlarge island (Fig.
1).
METHODS
Researchcommencedon 1 June.Geesehad already
begun egg laying [the earliestbeginning of a nesting
seasonrecorded for Ross' and Snow geese in this
area (cf. Ryder 1967, 1972)]. Clutch sizes were recordedduring nest surveysof the colony.These surveys (23 June-1July) were conductedduring the incubation period after all clutches were completed.
Egg size was usedto determinespecies(Ryder1971b)
when nesting birds were not identified by direct observation.Clutchesthat containedeggsof both Ross'
and Snow geesewere excludedfrom analyses.Habitat surrounding each nest site was partitioned into
five types,similarto thosedelineatedin Ryder'sstudy
(see Ryder 1972 for detailed descriptions).Briefly,
these included:
(1) "Rock habitat" was composedof gravel and
rocksof varioussizeswith almostno vegetationapart
from a few lichens.Rock habitat occurredalong the
tops of drumlins (glacialridges)and was free of snow
earlier than other habitats, probablybecauseof exposure to wind.
(2) "Mixed habitat" occurredalong lower elevation drumlinsand includeda scatteringof a variety
of low-growing vascularplants, lichens, and mosses
on a relatively dry, gravel-filled soil. This was the
most commonupland habitat in the colony.
(3) "Heath patches" occurred in wind sheltered
placesalong the slopesof rock ridgesand were characterized by Labrador tea (Ledurndecurnbens)and
white heather (Cassiopetetragona),as well as a few
dwarfbirch(Betulaglandulosa)
andwillow (Salixspp.).
Ryder (1972) included heath patches with mixed
habitat.
(4) "Sand tussock" was an uncommon habitat of
low-lying areascharacterizedby frost-heavedpingos
of sandthat were oftenringedby sedges(Carexspp.)
or other vascular plants.
(5) "Moss habitat" predominatedin all low-lying
areas of the colony and consistedof a wet mat of a
variety of mossesoverlying 5-10 cm of peat.
I establishedsix61-m-diametercircleplots(cf. Ryder 1969b)and six 305- x 6-m strip plots (4-19 June)
on island and mainland sites of the colony during
April 1983]
TABLE 1.
337
Ross'and SnowGooseNesting
Goose nest densities in different
habitats at Karrak Lake in 1976.
Nests/i,000 m 2 (tests)a
Samplearea(m2)
Habitat (code)
Rock (A)
Heath (B)
Mixed (C)
Tussocks (D)
Moss (E)
Islandt'
Mainland t'
2,586
1,994
9,522
166
3,462
0
0
2,629
2,634
0
Ross'Geese
Island
SnowGeese
Island
Mainland
16 (BCD) a
64 (ACE)
40 (ABE) P < 0.01½ 3 (D)
42 (AE)
ns
23 (C)
18 (BCD)
23 (BCE)
14 (AE)
14 (AE)
6 (0)
Mainland
nsc
ns
16 (0)
11 (0)
_2(ABC)
Habitats (codes)with statisticallydifferent nest densities(P <: 0.05, x2-tests);0 - none different.
Line plots(6) and circleplots(4) on islands;circleplots (2) on the mainland(0 indicatesno habitatin sampleplots).
Differencesbetween adjacentvalues, x•-tests;ns = not significant.
determined from 11 circle plots measured in
1968by Ryder(1969b)(t = 3.19,P < 0.05). Nest
densitiesvaried greatlyamonghabitats (Table
the egg-laying period in order to determine nest
densitiesand hatching success.I constructedscale
maps of these plots and determined nest locations
by "on-site" measurement to the nearest 15 cm.
Habitat typeswere delineatedwithin eachplot, and
areas were determined from scale maps with a planimeter.
Distances
1). Numbers of nests observed in the various
habitatswere significantlydifferentfrom numbers that would be expectedif nest sites were
distributed independently of habitat types
of a nest to the next nearest nest
and nearest conspecificnest were measured from
maps of circle plots (nests that were closer to plot
borders than their nearest neighborswere excluded
from analyses).I recordedthe clutchsize of all nests
locatedin circleand strip plots during the incubation
period (including nests already abandoned).Hatching success
was determinedwhen nestswere checked
after hatching (9-13 July). I used standard statistical
methods, except for the analysesof clutch size, for
which I used t-tests for unequal variances in test
populations (Remington and Schork 1970: 212) becauseBartlett'stestsfor homogeneity(Snedecorand
Cochran1967:297) revealedsignificantdifferencesin
variance among test populations.
(Ross', X•= 212.7, df = 6, P < 0.001; Snow,
X• = 54.7, df = 5, P < 0.001). Highest densities of nesting Ross' Geese occurredin heath
patcheson islandsthat had been occupiedby
geesein the late 1960's, and they were fourfold greater than the lowest densitiesin rock
and mosshabitats. Highest densitiesof Snow
Geeseoccurredin rock habitat on glacialridge
tops, and relatively dispersednesting occurred
in mosshabitat. BecauseRyder (1969b, 1972)
and I probably classifiedhabitats differently, I
did not compare our results in detail.
Topographicand habitat featuresof the largest island
RESULTS
at Karrak
Lake were more similar
to
mainlandareasthan they were to smallislands
occupiedby nestinggeese(seealsoRyder1972).
Clutch sizesof both Ross'and Snow geeseon
this islandwerenot significantlydifferentfrom
those of geesenesting in comparablehabitats
Overall nest densities in circleand strip plots
(which included snow-covered areas that were
unavailablefor nesting) averaged36/1,000m2,
significantlygreaterthan the 12 nests/i,000m2
TABLE2. ROSS'Gooseclutch-sizevariation among island and "expansion" area habitats at Karrak Lake in
1976.
Clutch size (• + SE)
Habitat (code)
Rock (A)
Heath (B)
Mixed (C)
Tussocks(D)
Moss (E)
Islands
3.3
3.6
3.5
3.5
3.2
+
+
+
+
+
0.12 (BC)a
0.06 (AE)
0.03 (AE)
0.12 (E)
0.10 (DCB)
n
78
353
1,106
54
110
Expansion
nst,
P < 0.05b
P < 0.10
ns
Habitats with statisticallydifferent dutch size (P < 0.05, t-tests);0 = none different.
Differencesbetweenadjacentmeans,t-tests;ns = not significant.
3.1 + 0.26 (0)a
no data
3.3 + 0.04 (0)
3.3 + 0.05 (0)
3.3 + 0.22 (0)
n
9
483
399
12
338
M. ROBERTMcLAtaDRESS
TABLE 3.
[Auk, Vol. 100
Snow Goose clutch-size variation among island and "expansion" area habitats at Karrak Lake in
1976.
Clutch size (• + SE)
Habitat (code)
Rock (A)
Heath (B)
Mixed (C)
Tussocks(D)
Moss (E)
Islands
4.2
4.5
4.4
4.4
3.8
+
+
+
+
+
0.10 (0)a
0.17 (0)
0.05 (0)
0.41 (0)
0.40 (0)
n
146
68
530
9
9
Expansion
n
P < 0.10a 4.0 + 0.12 (E)
no data
P < 0.05 4.0 + 0.05 (E)
ns
4.0 + 0.09 (E)
ns
3.4 + 0.22 (ACD)
153
918
130
28
a AS in Table 2.
at mainland sites. Therefore, clutch-size data
nificantly with distance from nests of Snow
from the largeislandand mainland(referredto Geese. At distances of less than 3 m, nearest
as "expansion"sites)were pooledfor further neighbors of Snow Geese were Ross' Geese
more than twice as often as they were Snow
analyses.
The averageclutchsize of Ross'Geesedif- Geese, despite a near equal overall occurrence
feredsignificantlyamonghabitats(Table2). On of either speciesasnearestneighbors(Table 4).
islands,clutcheswere 10% largerin high-den- Speciescompositionof nearestneighborsdid
sity nestsin heath patchesand intermediate- not change significantly with distance from
densitynestsin mixed and sandtussockhab- Ross'Goosenests.Also, as nearestneighbors,
itats than were clutchesin low-density nestsin twice as many Ross'Geesenestedless than 3
rock and moss habitats. Mean clutch sizes of
m away than at distancesexceeding3 m from
Ross' Geese nesting in expansion sites were
both Ross' and Snow goose nests. From these
similar to averageclutchesin island habitats data it appears that Ross' Goose responsesto
where nest densities were low. Clutch sizes of
Snow Geeseas nearestneighborswere not dif-
Snow Geese were not statisticallydifferent ferent from responsesto its own speciesfor
amongislandhabitats(Table3). Snow Geese purposesof spacing.
No clear patterns between clutch size and
that occupiedrock and mixed habitats in expansion areas, however, had significantly nearest nest distance were evident when
smaller clutchesthan island-nestingSnows in neighborswere consideredwithout respectto
the same habitats.
species(seealsoRyder1969b,1972).Therewas
Nearest neighborswere conspecificsmore an inverse correlation between clutch size of
frequentlythanwould be expectedif Snowand successfulnests and nearest conspecificnest
Ross'geesewere distributedthroughoutthe distance for Snow Geese, however (Fig. 2).
colonyindependentlyof one another(X•= Successfulnestswith largerclutcheshad closer
26.25, df = 1, P = 0.001; Table 4). Species conspecific
neighborsthan did nestswith small
compositionof nearestneighborschangedsig- clutches. Ross' Geese, on the other hand, did
TABLE4. Relationshipbetween speciesof nearestneighbor and intemest distanceof Ross'and Snow geese
in circle plots at Karrak Lake in 1976.
Species/number
of nests
Snow Geese
n 146
=
Ross' Geese
n=310
•P
< 0.001.
Not significant.
Number
ofnearest
neighbors
that
were:
Distanceto nearestneighbor
Snow
Less than 3 m
24
6mormore
44
Total
68 (47%)
Less than 3 m
6mormore
40
Total
71 (23%)
31
Ross'
(X
2=13.17)
a 51
27
78 (53%)
(X
2=2.26)b
158
81
239 (77%)
April 1983]
Ross'andSnowGoose
Nesting
339
TanrE 5. Relationship between speciesof nearest
neighbor and nesting successof Ross' and Snow
geeseat Karrak Lake in 1976.
SNOW
Percentage(n) of nests
successful
F=1.36, P:'"0.1
Speciesof
nearest neighbor
r=0.95, P"c0.01
b.I
Ross'
Snow
Z
43
ROSS'
bJ
Z
•
91 95 33
Snow Geese and 0 Ross' Geese), as were nests
F=1.07, P:,"0.1
Z
O
o
2'
I
1,2
81 (78)
76 (68)
largeclutches(>8 eggs)were extremelyrare (3
29
(,/3
80 (239)
80 (71)
by Snow Geese without their own nests is
common in some years and results in higher
than averagerecorded clutch sizes (Syroechkovsky 1979).Overall averageclutchsizeswere
not high (cf. Owen 1980), however, and very
-1-
(_3
bJ
Snow Geese
27
r•
0
(.•
Ross' Geese
I
I
I
I
3
4
5
6
containingeggsof both species(n = 5, 0.1%
of 4,600 nests).Egg losswas not a major factor
affectingaverageclutchsizeseither.Of 379eggs
(94 clutches)that were marked(6 June)during
the egg-layingperiod, only 6 eggs(2%) were
lost before the last day of nest surveys on 1
July. Even this low rate of egg loss probably
exceeded losses from unmarked nests, because
CLUTCH
SIZE
Fig. 2. Relationshipbetween clutchsize and distance to nearestconspecificnest (œ+ 1 SE) among
successfulRoss' and Lesser Snow geeseat Karrak
Lakein 1976.Numbersbeneathbarsaresamplesizes.
the disturbanceof repeated nest checksmay
have contributed to partial clutchloss to avian
predators(Macinnes 1980, Strang 1980).
Colonyexpansion.--InSnow Geese,older and
earlier-nestingfemaleslay the largestclutches
(Finhey and Cooke 1978), and they tend to be
traditional in their nest location (Cooke and
not exhibit a statisticallysignificantrelationship, although six-egg nests tended to have
closerconspecific
neighborsthan did nestswith
smallerclutches.Overall nestsuccess
was high
(Ross': 84%, n = 692; Snow: 81%, n = 333; ex-
cludingone-eggnests;cf. Ryder 1971a,1972).
Nest success
did not vary significantlyrelative
to the speciesof nearestneighbor (Table 5).
DISCUSSION
Abraham 1980). Thus, many Snow and Ross'
geesenestingin the densestisland areasof the
colony were probably older birds, which initiated nesting early in the season.The recent
expansionof the colony to the largest island
and mainland areas, coupled with lower average clutchsizes at theselocations,probably
indicates that a higher proportion of young
geeseoccurredamongbirds nestingin expansion areas than among island nesting geese
(Tables 2 and 3). Recall, however, that main-
Variation in averageclutchsizesof Ross'and
Snow geese among habitats (Tables 2 and 3)
was most likely attributable to corresponding
differencesin ageand/ortime of layingof nesting geese(seebelow). Two other factors,nest
parasitism (Mineau and Cooke 1979) and egg
lossduring the period of egglaying and early
incubation, may have affected the recorded
clutchsize of somenests.Parasiticegg laying
land nesting Snows had attained densities
similar to or even greater than Snow Geese on
islandsin the samehabitat type (Table 1). Presumably,young Ross'Geeseutilized the habitats of low-lying areas(mossand tussocks)on
islandsto a greaterextentthan did SnowGeese
and were presentonly in low densitieson the
mainland (relative to similar island habitat).
Thesepatternssuggestthat younggeesewere
340
M. ROBERT
McLAI•IDRESS
[Auk, Vol. 100
probably young or late nesters,based on their
smallclutches.In other habitats,vegetationwas
generallyusedfor nestconstruction,especially
habitats.
for nestsin heath patch and mosshabitats (see
Maximum
nest densities
attainable
in a
also Ryder 1967). These areas were occupied
mixed-speciescolony of Ross'and Snow geese predominantlyby Ross'Geese.Unmelted snow
are difficult to predict. Densities of nesting was common in heath patches and moss habgeesereported in other studies are only qual- itat in early June, indicating that many nest
itatively comparableto those at Karrak Lake sitesin thesehabitatswere unavailableto geese
becauseof differences in methodologiesused arriving beforethe studybeganon 1 June.Perand nonuniformity of habitat. Typically, the haps availability of nest material is an imporhighest concentrationsof Lesser Snow Goose tant proximate factor in the selection of nest
nestsreported at other coloniesvary from 2 to sites by Ross' Geese. Becauseof their smaller
6 nests/I,000 m• (Uspenski 1965, Barry 1967, body size, Ross' Geese would be expected to
Harvey 1971, Ryder 1972, Kerbes 1975, Finney expendrelatively more energyfor maintenance
and Cooke1978;my calculations).Cooch(1958), during the incubation period and should benhowever, reportedterritoriesof 15-32 m2/nest, efit more from a favorable nest microclimate
indicatingdensitiesof 30•6 nests/I,000m2, and than would Snow Geese.
found someterritoriesof lessthan 5 m2. Ryder
An important aspectof nest-siteselectionin
(1967)reporteddensitiesof more than 200 Ross' colonially nesting birds is safety from predaGoosenests/I,000m2 in a colonyat ArloneLake, tion (Buckleyand Buckley1980).Predation was
approximately50 km west of KarrakLake. It is not a major factor affecting goosenest success
possible then, that in the absenceof suitable in 1976, however. Few avian predators were
habitat on mainland areas, Snow and/or Ross' present during early incubation, probably begeese might have achieved even higher nest cause of inclement weather from 1 to 15 June.
densities at Karrak Lake.
Apart from a few Long-tailed Jaegers(StercoNest spacing and habitat selection.--Snow rariuslongicaudus),
avian predatorsdid not nest
Geese tend to arrive at coloniesa day or two successfully.
This probablyresultedin low food
earlier and begin laying eggs sooner than do requirements relative to "normal" years. Also,
Ross' Geese (Ryder 1967). Snow Geese ap- northern red-backed voles (Clethrionomysrupeared to space themselves relative to one tilus)were visibly abundantin earlyJune.This
another and may have ignored Ross'Geeseor, provided a food sourceother than gooseeggs
at least, not reacted to them in the same manfor both avian and mammalian predators.
ner as their own speciesfor purposesof spac- Often, goose nests were abandoned several
ing. Ross' nested closer to Snow Geese than days before being scavengedby "predators."
did other Snow Geese (Table 4), but nest suc- Nevertheless,predation has been a significant
cesswas not affectedby the speciesof nearest factor affecting nest successat Karrak Lake in
neighbor (Table 5).
the past (Ryder 1972).
Habitat selection by both Ross' and Snow
Predation may also be an important factor
geesewas indicated by high nest densitiesand influencing nest-site selectionby geesewithin
correspondinglyhigh clutchsizesin preferred a colony. Ryder (1972) proposed that Ross'
habitats. Clutch size and nest densities of Ross'
Geese select islands for nesting becausethey
Geese were highest in heath patches.A pref- provide safety from predation by arctic foxes
erence by Ross' Geese for nest sites "next to (Alopexlagopus).Ross' Geese are agile flyers
rocks or birch patches," reported by Ryder and adept at chasingjaegers(pers.obs.). Snow
(1972: 193), indicated that geeseselectedhab- Geese, on the other hand, are capable of deitat features in the late 1960's that were most
fending against arctic foxes (Barry 1964, Syrcommon in heath patches in 1976. In 1976, oechkovsky1972)and commonlynest on mainhighest nest densities of Snow Geese were in land areas at other arctic colonies. Thus,
the earliest snow-free areasalong the tops of expansionof the Karrak Lake colonyto include
the highest drumlins. For both species,nests nearby mainland areas probably began with
in rock habitat were little more than scraped pioneering by Snow Geese, which predomidepressionsin the gravel, but few Ross'Geese nated in mainland nesting areas.
nested in rock habitat, and those that did were
Nesting territoriesof colonialgeese.--Ryder
unable to compete with older geese of either
speciesfor nest sites and were forced to nest
in mainland and less denselyoccupiedisland
April 1983]
Ross'andSnowGooseNesting
341
TABLE6. Changein nearestneighbordistancesof Ross'Goosenestsat Karrak Lake between 1966-68and
1976.
Distance (m) to nearestnest (• + SE)a
Year
1966-1968
Successful nests
n
4.7 + 0.1b
413
P < 0.001
3.0 + 0.1
1976
249
Unsuccessful nests
P • 0.01e
ns
4.1 + 0.2b
P < 0.001
2.7 + 0.1
n
114
59
NeighborswereeitherRoss'or LesserSnowgeese.
Source:Ryder(1972).
Differencebetweenadjacentmeans,t-tests;ns = not significant.
(1967, 1975)and Inglis (1976)suggestedthat dation on Snow Goose nests by foxes on
foodwas an importantattributeof nestingter- Wrangel Island, Siberia was least where nestritoriesof colonialgeese.This hypothesiscan- ing density was highest (see also Cody 1971).
not be discounted for Ross' Goose territories
At Karrak Lake, Ross' Geesepursued jaegers
despite their small size, becausefood (vege- well beyond the boundariesof their nesting
tation) may be more concentratedin optimal territories.Thus, colonialgeeseshouldbenefit
nesting areas (e.g. the heath patches). Snow (perhapsinadvertently)from nest defenseby
Geese,however,nestedat highestdensitiesin neighbors in high-density areas. If so, selecrockhabitat,whichwasalmostdevoidof vege- tion should favor individuals that tolerate close
tation. In addition, Snow Geese should defend neighbors. Intraspecificaggressionat nesting
food resources
from neighboringRoss'as well areas is often interpreted as defense of a "teras neighboringSnows,as it is unlikely that ritow" in geese. Aggressionamong colonial
food is in sufficient abundance for the two
geese,however, may not be related solely to
speciesto eatdifferentfoodswhennesting.In- maintenance of a specific area. Mineau and
terspecific nearest nest distanceswere smaller Cooke (1979) argued that male Snow Geese
than nearestneighbordistancesbetweenSnow "protecttheirparenthood"
andguardonlytheir
Geese(Table4), whichsuggests
thatRoss'may mates (from rape) and nests (from parasitism).
be toleratedwithin intraspecificterritoriesof In addition, or alternatively, thesebehavioral
Snow
Geese.
interactions may be important to assure that
Hypothesesconcerningterritorialityin co- conspecific
neighborsarein goodconditionand
lonialgeese(Ryder1975,Inglis1976,Owen and remain attentive to their nests, thereby inWells1979,MineauandCooke1979)imply that creasingthe probability of reciprocalnest debirds in the smallest territories would have less
fense (see discussionof Wasser 1982). Old, ex-
food or more intraspecificstrife (rape and/or
conflict)than birdsin the largestterritories.In
Ross'Geese,smallterritorieswere previously
thoughtto decreasenestingsuccess,because
periencedbirds would be ideal neighborsand
expectedto predominatein the densestnesting
areas. If young and/or late arriving birds are
preventedfrom nestingin establishedareasof
the colony,as suggestedfrom clutch-sizedis-
unsuccessful
birds
tended
to have
closer
neighborsthan successful
birds (Ryder1972). tributions at Karrak Lake, then predation
A comparison of nearest-neighbordata ob- shouldbe reducedby enhancement
of reprotainedfrom1966to 1968(Ryder1972)with sim- ductive synchrony (Darling 1938, Gochfeld
ilar data obtainedin 1976(this study)shows 1980).
Competition.--MacInnes and Cooch (1963)
that successful
nestshad closerneighborsin
1976 than either successful or unsuccessful nests
suggestedcompetitionwith Snow Geesemay
10 yr earlier (Table 6). Despite the decreasein have led to the restrictedbreedingdistribution
spacing,overallnestingsuccess
was high (see of Ross'Geese.Ross'do not avoidnestingin
above).
colonies with Snow Geese, however, and, at
For colonial nesting to be successful,any
disadvantagesof small territoriesmust be offsetby advantagesof having conspecifics
nearby. Syroechkovsky(1972) reported that pre-
Karrak Lake and other large coloniesin the
centralArctic, both specieshave been highly
successful in recent years (Kerbes et al.
1983). Further, there was no indication that
342
M. ROBERT
MCLANDRESS
[Auk, Vol. 100
convergentevolution in seabirdcommunities.A
either specieswas adverselyaffected by the
comment. Ecology57: 177-184.
other's presencein this study (see also Ryder
BELLROSE,
F. C. 1980. Ducks, geeseand swans of
1967). Ross'had a high tolerancefor crowding
and nested
close to Snows.
Snow
Geese
were
able to nest in areas that Ross' avoided, such
as on the mainland or on rock ridges on is-
North America. Harrisburg,
StackpoleBooks.
BUCKLEY,F.G.,
Pennsylvania,
& P. A. BUCKLEY. 1980. Habitat
selectionand marine birds. Pp. 69-111 in Behavior of marine animals: current perspectives
in research,Vol. 4: marine birds (J. Burger, B.
lands. The two species have different, but
complementary, antipredator behavior. This
L. Olla, and H. E. Winn, Eds.). New York, Pleprobably accountsfor the ability of Ross' to
num Press.
nest successfully
on mainland areaswhen Snow
BURGER,
J. 1981. A model for the evolution of mixedGeese are present for defense of nests against
species
colonies of Ciconiiformes. Quart. Rev.
foxes.Reciprocally,closeRoss'neighborsmay
Biol. 56: 143-167.
benefit Snow Geese by protectingtheir nests
COD¾,M. L. 1971. Ecologicalaspectsof reproduction. Pp. 461-512 in Avian biology (D. S. Farner
andJ.R. King, Eds.).New York,AcademicPress.
from avian predators.
If competitionwith Snow Geeseaffectsthe
breeding distribution of Ross'Geese,it prob1973. Coexistence, coevolution and conably occursindirectly. For example,interaction
vergent evolution in seabird communities.Ecolat post-hatchfeeding sitesor on migration and/
ogy 54: 31-44.
or wintering areasmay haveled to the restrict- COOCH,F. G. 1958. The breeding biologyand management of the Blue Goose(Chencaerulescens).
ed range of Ross'Geese.Two sympatricspecies
Unpublished Ph.D. dissertation, Ithaca, New
as closely related as Ross' and Snow geese,
however, would be expected to avoid compe-
tition and probablydiffer in their use of limited resources(see Lack 1971), at least in areas
unalteredby man.
ACKNOWLEDGMENTS
York, Cornell Univ.
COOKE, F., & K. F. ABRAHAM. 1980. Habitat and
locality selectionin LesserSnow Geese:the role
of previous experience. Proc. 17th Intern. Ornithol. Congr.: 998-1004.
DARLING,F. F. 1938. Bird flocksand the breeding
cycle.Cambridge, Cambridge Univ. Press.
DELACOUR,J. 1954. The waterfowl of the world.
This study was funded primarily by Canadian
Wildlife Service (CWS) contract #0SZ5-0262 to RIM
EcologyLtd. (RIM). I am gratefulto M.D. Bell of
London, Countrylife.
FINNEY,G., & F. COOKE. 1978. Reproductivehabits
in the Snow Goose:the influenceof female age.
Condor 80: 147-158.
Winnipeg; I. McLandress and W. A. McLandress
(RIM); and T. W. Barry, A. Dzubin, and J. Patterson GOCHEELD,M. 1980. Mechanisms and adaptive
(CWS)for helpwith logisticproblems.Specialthanks
value of reproductivesynchronyin colonialseaare due G. W. Beyersbergen(CWS)for technicaladbirds. Pp. 207-270 in Behavior of marine anivice and help in gatheringfield data. S. Bartlettand
mals: current perspectivesin research,Vol. 4:
A. Crickmore(Universityof Californiaat Davis-UCD)
marine birds (J. Burger, B. L. Olla, and H. E.
provided valuable assistancewith data analysis.I
Winn, Eds.). New York, Plenum Press.
benefited from discussions with A. Dzubin, T. W.
HARVEY,J. M. 1971. FactorsaffectingBlue Goose
Barry,R. H. Kerbes(CWS);T. W. Aldrich,S.C. Minnesting success.Can. J. Zool. 49: 223-234.
ta, J. S. Sedinger(UCD); and J.P. Ryder (Lakehead INGLIS,I. R. 1976. Agonisticbehaviourof breeding
University). D.G. Raveling,D. W. Anderson,and
Pink-footed Geesewith referenceto Ryder's hyC. A. Toft (UCD) provided many valuable suggespothesis.Wildfowl 27: 95-99.
tionsforrevisionof earlydrafts.Reviewsby F. Cooke, JOHNSGIRD,P. g. 1978. Ducks, geese, and swans
C. Findlay, and M. Owen were also helpful.
of the world. Lincoln, Nebraska, Univ. Nebraska Press.
LITERATURE
CITED
BARRY,T. W. 1964. Brant, Ross' Goose, and Emperor Goose. Pp. 145-154 in Waterfowl tomorrow (J.P. Linduska,Ed.). Washington,D.C., U.S.
GovernmentPrinting Office.
1967. The geeseof the AndersonRiver Delta, Northwest Territories. Unpublished Ph.D.
dissertation, Edmonton, Alberta, Univ. Alberta.
BEDARD,J. 1976. Coexistence, coevolution and
KERBES,R. H. 1975. The nesting population of
LesserSnow Geese in the easternArctic: a photographicinventory of June, 1973. Can. Wildl.
Serv. Rept. Ser. 35.
, M. R. MCLANDRESS,G. E. J. SMITH, G. W.
BEYERSBERGEN,
& B. GODWIN. 1983. Ross' Goose
and Lesser
Snow
Goose
colonies
in the central
Canadian Arctic. Can. J. Zool. 61: 168-173.
LACK,D. 1971. Ecologicalisolation in birds. Cambridge, Massachusetts,
Harvard Univ. Press.
April 1983]
Ross'andSnowGoose
Nesting
MACINNES, C. D.
1980.
Comment:
Observer-in-
ducedpredationis real. J. Wildl. Mgrnt. 44: 222-
343
Ph.D. dissertation, Saskatoon, Saskatchewan,
Univ.
224.
Saskatchewan.
1971a. Distribution and breeding biology
, & F. G. COOCH. 1963. Additional
eastern
recordsof Ross'Goose (Chenrossii).Auk 80: 77-
of the Lesser Snow Goose in central Arctic
ada. Wildfowl
79.
1971b.
McLANDRESS, M.R.
1979. Status of Ross' Geese in
California. Pp. 255-265 in Managementand biologyof PacificFlyway geese(R. L. Jarvisand J.
C. Bartonek, Eds.). Corvallis, Oregon, OSU
Can-
22: 18-28.
Size differences
between
Ross' and
Snow Goose eggs at Karrak Lake, Northwest
Territories
in 1968. Wilson
Bull. 83: 438-439.
1972. Biology of nesting Ross'sGeese. Ardea 60: 185-215.
Bookstores, Inc.
1975. The significanceof territory size in
MINEAU, P., & F. COOKE. 1979. Territoriality in
colonialnestinggeesedanhypothesis.Wildfowl
26: 114-116.
Snow Geese or the protection of parenthood-Ryder's and Inglis's hypothesis re-assessed. SNEDECOR,G. W., & W. G. COCHRAN. 1967. StaWildfowl
30: 16-19.
OWEN, M. 1980. Wild geeseof the world. Norfolk,
England, B. T. BatsfordLtd.
, & R. WELLS. 1979. Territorial behaviour in
breeding geese•a re-examinationof Ryder's
hypothesis. Wildfowl 30: 20-26.
REMINGTON, g. D., & M. A. SCHORK. 1970. Statis-
tistical methods. Ames, Iowa, Iowa State Univ.
Press.
STRING, C. A. 1980. Incidence of avian predators
near people searching for waterfowl nests. J.
Wildl. Mgmt. 44: 220-222.
SYROECH•COVS•CY,
E.V. 1972. [Some peculiarities of
interrelationships between the Snow Geese and
ticswith applicationsto the biologicaland health
the arcticfoxeson the WrangelIsland.]Zool. Zh.
sciences.EnglewoodCliffs, New Jersey,Pren-
51: 1208-1213. (In Russian.)
tice-Hall
1979. [The laying of eggs by white geese
into strange nests.] Zool. Zh. 58: 1033-1041. (In
Russian.)
Inc.
RYDER,J. P. 1967. The breeding biology of Ross'
Goosein the PerryRiver region,NorthwestTerritories.Ottawa, Ontario. Can. Wildl. Serv. Rept.
Ser. No.
3.
1969a. Nestingcoloniesof Ross'Goose.Auk
86: 282-292.
1969b. Timing and spacing of nests and
breeding biology of Ross'Goose.Unpublished
USPENS•C•,
S.M. 1965. The geeseof WrangelIsland.
Wildfowl Trust Ann. Rept. (1963-1964)16: 126129.
WASSER,S. K. 1982. Reciprocity and the trade-off
betweenassociate
quality and relatedness.
Amer.
Natur.
119: 720-731.