Use care in determining age-ratios in shorebirds: they may differ
relative to flock position, flock location and behaviour
BRIAN
HARRINGTON
ManometCenterfor ConservationSciences,Manomet,MA USA 02345. [email protected]
Harrington,B. 2004.Usecarein determiningage-ratios
in shorebirds:
theymay differrelativeto flockposition,
flock locationandbehaviour.Wader StudyGroup Bull. 104: 92-93.
I reportvariations
in age-ratios
withina flockof RedKnotsandbetweenflocksof Sandealings.
Theseobservations
haveimportantimplications
for thoseattemptingto measuretheproportionof juvenilesin wholepopulations.
INTRODUCTION
Sanderling
Sampling the proportion of juveniles among flocks of
shorebirds
duringsouthwardmigrationis a potentialtoolfor
gauginghow successfulannualreproductionhas beenfor
Arctic-breedingand other shorebirds.This note describes
potentialsamplingbiasthatcanoccurdueto the non-random
distributionof juvenile and older/adultshorebirdsin roosting flocks.AlthoughI am awareof thisissuein a varietyof
shorebirdspecies,for simplicity'ssakeI will focuson two
examples:a single group of Red Knots Calidris canutus
roostingat a migrationstopoversite on the Georgia(Atlantic) coastof the United Statesandseparateflocksof Sanderlings C. alba usinga singleMassachusetts
beach.
METHODS
The observationsof juvenile and older Sanderlings(see
Haymanet al. 1986for plumagedescriptions)
weremadeon
PlymouthBeach,Massachusetts,
a peninsularbarrierbeach
extending about 5 km into Plymouth Harbor (42ø05'N,
70ø40'W, see Schneider& Harrington 1981 for map). All
countswere madewithin 1.5 hr of high tide on 22 September2003. At hightidemostshorebirds
ceaseforagingandfly
to rooston nearbybeaches.All Sanderlingson the beach
were counted,but were foundin only four of 14 surveysections.I sampledtwo of thesesectionsagainon 26 September, making a full countin one and samplingfrom larger
numbersin the second,but taking care to ensureas best I
couldthatthe sampleswererepresentative
by countingfrom
differentpartsof the flock.
Red Knot
RESULTS
Observations
weremadeat Little St. SimonsIsland,Georgia,
USA (31 o17'N, 81o17'W), on 12 September2003 on a flock
of about 2,250 Red Knots. They began an hour before a
1030h high tide, and continuedfor the following 2.5 hrs.
Initially the knotswere in a singleflock, but after an hour,
oceanwaves washedinto the roostinglocationand the sin-
Knots in Georgia
Theproportionofjuveniles(Table1) ontheedgeof thelarge
flock was significantlyhigherthanthe interiorof the flock
(Z2 = 13.47,p < 0.01),andsignificantly
higherthanfound
in the'random'
sample
(Z2= 8.03,p < 0.01).Theproportion
gleflocksplitintothreegroups
of roughlyequivalent
size, of juvenilesin the interiorof the flock wasnot significantly
one located close to the crest of the beach and the others
different(Z2= 3.12,p > 0.05)fromtheproportion
foundin
roughly 15-30 m furtherwest (inland).
Basedon plumagecharacteristics
(Hayman et al. 1986),
the age of eachknot was determinedasjuvenile or older.
While groupedas a singleflock, I countedjuvenile: older
ratios(a) nearthe centreof the flock, (b) at the edgeof the
flock, and(c) at 12 'random'pointlocationswithin theflock
(with a goal of 30 birdsat eachpoint) alongtwo imaginary
transects,one runningthe lengthand the other runningthe
width of the flock (the countat one point was incomplete
becausethe flock flew).
About 30 minutesafter the flock split into threegroups,
individualandsmallgroupsof knotsbeganto walk from the
two westerlyflocks to join with the flock at the crestof the
beach.I countedageratiosfromthesebirdsfor aboutanhour
astheywalkedbetweentheflocks;thisendedwhena raptor
the 'random'sample,a resultthatis not surprisinggiventhat
a vast majority of the knotswere in the interior versusthe
edgeof the flock.
When, after the single,large flock split into three,I tallied the ageratiosof birdswalkingbetweenflocks,the proportion of juveniles was lower than in any of the previous
samples(Table 1) andsignificantlylower thanfoundin the
flushed all the birds.
I found Sanderlingsin four of my surveysectionson Ply-
Bulletin 104 August 2004
92
'random'sample(Z2 = 24.32,p < 0.01). Shortlybeforea
raptormadeall thebirdsfly, I noticedthattherelativelysmall
flocks remainingto the west (from which the walking birds
had beendeparting)had a high proportionof juveniles,but
I did not obtaincountsbeforethe raptorappeared.
Sanderlings in Massachusetts
93
Harrington:Use care in determiningage-ratios of shorebirds
Table 1. Differingage ratios of Red Knots relative to positionin a
flook and to behaviour.
Juveniles
Older
than
juveniles
Interior
of flock
Edge of flock
'Random' location (see text)
Walking betweenflocks
4
17
24
27
137
83
313
1,353
Table 2. Uneven age ratios of Sanderlingsin differentflocks on the
same beach, 22 September 2003.
Per cent
Juveniles
juveniles
Older
than
Per cent
juvenile
juvenile
3
South end of beach
3
0
100
17
7
2
North end of beach
17
1
94
Mid-north
55
mouth Beach on 22 Septemberand in three of theseall but
a couplewerejuveniles;in theremainingsectionalmost90%
wereolderbirds(Table2). I wasnotawareof anymovement
betweenthe sections.On 26 September,two of thesesections
contained respectively 95% juveniles (N - 52) and 8%
juveniles (N = 203).
DISCUSSION
This studydemonstrates
thatadultandjuvenile shorebirds
do
not necessarilysort randomlyin high tide roostingflocks.
This hasimportantimplicationsfor thoseattemptingto measure the proportionof juveniles in a populationin order, for
example, to estimaterecruitment(see Clark et al., this vol-
section
South of mid-north
section
425
34
11
1
97
ACKNOWLEDGEMENTS
I am gratefulfor logisticsupportfrom naturalistsand other
staffof The Lodgeat Little St. SimonsIsland,Georgia,from
Brad Winn, GeorgiaDepartmentof Natural Resources,and
from the Manomet
Center for Conservation
Sciences. I thank
JacquieClark for commentson the draftmanuscript.Counts
of Sanderlings
were madeas a part of work funded,in part,
througha grantfrom the Delaware CoastalProgramswith
fundingfrom the Office of OceanandCoastalResourceManagement,NationalOceanicandAtmosphericAdministration
under award numbers NA17OZ2329
and NA17OR2479.
REFERENCES
ume).
A numberof studieshaveshownthat age-ratiosof shorebirdscan vary with factorssuchasroostlocation(Swennen
1984,Durrell et al. 1996), vulnerabilityto raptors(Cresswell
1994), or foraginghabitat(Wamock& Takekawa1995). My
observationsshow that juvenile Red Knots were not randomly dispersedwithin a single roosting flock, and that
juvenile Sanderlingswere not randomlydispersedbetween
roostingflocks.This suggests
that attemptsto estimateageratiosof a wholepopulationneedto be doneusingasrandom
and representativea samplingprocessas possible.Ideally,
samplesshouldbe takenfrom a wide rangeof sitesacross
entirelandscapes.
Further,theresultssuggestthat age-ratios
can vary with differing behaviour,as for example among
birds switchingflocks.
Cresswell,W. 1994. Age-dependentchoiceof Redshank(Tringa totanus)
feeding location- profitability or risk. J. Anim. Ecol. 63: 589-600.
Durrell, S.E.A. le V dit, S.J. Ormerod & P.J. Dare. 1996. Differences
in population structure between two OystercatcherHaematopus
ostralegusroostson the Burry Inlet, SouthWales. Ardea 82A: 283388.
Hayman, P., J. Marchant & T. Prater. 1986. Shorebirds:an identification guide to the waders of the world. HoughtonMifflin, Boston.
Schneider,D.C. & B.A. Itarrington. 1981. Timing of shorebirdmigration in relationto prey depletion.Auk 98: 197-220.
Swennen,C. 1984. Differencesin quality of roostingflocks of Oystercatchers.pp. 160-176, In P.R. Evanset al. (eds). Coastalwadersand
wildfowl in winter. CambridgeUniversityPress,Cambridge.
Warnock, S.E. & J.Y. Takekawa. 1995. Habitat preferenceof wintering shorebirdsin a temporallychangingenvironment:WesternSandpipersin the San FransiscoBay estuary.Auk 112: 920-930.
Bulletin 104 August 2004