ArticlePDF Available

Using ring-recovery and within-season recapture data to estimate fecundity and population growth

Authors:

Abstract

Tag‐recovery data from organisms captured and marked post breeding are commonly used to estimate juvenile and adult survival. If annual fecundity could also be estimated, tagging studies such as European and North American bird‐ringing schemes could provide all parameters needed to estimate population growth. I modified existing tag‐recovery models to allow estimation of annual fecundity using age composition and recapture probabilities obtained during routine banding operations of northern pintails (Anas acuta) and dark‐eyed juncos (Junco hyemalis), and I conducted simulations to assess estimator performance in relation to sample size. For pintails, population growth rate from band‐recovery data (λ = 0.93, SD: 0.06) was similar but less precise than count‐based estimates from the Waterfowl Breeding Pair and Habitat Survey (λ: 0.945, SE: 0.001). Models with temporal variation in vital rates indicated that annual population growth in pintails was driven primarily by variation in fecundity. Juncos had lower survival but greater fecundity, and their estimated population growth rate (λ: 1.01, SD: 0.19) was consistent with count‐based surveys (λ: 0.986). Simulations indicated that reliable (CV < 0.10) estimates of fecundity could be obtained with >1,000 within‐season live encounters. Although precision of survival estimates depended primarily on numbers of adult recoveries, estimates of fecundity and population growth were most sensitive to total number of live encounters. Synthesis and applications: Large‐scale ring‐recovery programs could be used to estimate annual fecundity in many species of birds, but the approach requires better data curation, including accurate assessment of age, better reporting of banding totals, and greater emphasis on obtaining and reporting within‐season live encounters.
Ecology and Evolution . 2018 ;1– 8 .    
|
 1
www.ecolevol.org
Received:27Januar y2018 
|
  Revised:27J uly2018 
|
  Accepted:7August2018
DOI: 10.100 2/ece3.450 6
ORIGINAL RESEARCH
Using ring- recovery and within- season recapture data to
estimate fecundity and population growth
Todd W. Arnold
Thisisanop enaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium,
providedtheoriginalworkisproperlycited.
©2018TheAuthors.Ecology and EvolutionpublishedbyJohnWiley&SonsLtd.
Depar tmentofFisheries,Wildlife,and
Conser vationBiolog y,Universityof
Minnesota,St.Paul,Minnesota
Correspondence
ToddW.Arnold,D epartmentofFisheries,
Wildlife,andConservationBiology,
UniversityofMinnesota,St.Paul,MN.
Email:arnol065@umn.edu
Abstract
Tag-recoverydatafromorganismscapturedandmarkedpostbreedingarecommonly
usedtoestimate juvenileandadult survival. Ifannualfecundity couldalsobeesti-
mated,taggingstudiessuchasEuropeanandNorthAmericanbird-ringingschemes
couldprovideallparametersneededtoestimatepopulationgrowth.Imodifiedexist-
ingtag-recoverymodelstoallowestimationofannualfecundityusingagecomposi-
tion and recapture probabilities obtained during routine banding operations of
northernpintails(Anas acuta)anddark-eyedjuncos(Junco hyemalis),andIconducted
simulations toassess estimatorperformanceinrelationtosamplesize.Forpintails,
populationgrowthratefromband-recoverydata(λ=0.93,SD:0.06)wassimilarbut
less precise than count-based estimates from the Waterfowl Breeding Pair and
HabitatSurvey(λ:0.945,SE:0.001).Modelswithtemporalvariationinvitalratesin-
dicatedthatannualpopulationgrowthinpintailswasdrivenprimarilybyvariationin
fecundity.Juncoshadlowersurvivalbutgreaterfecundity,andtheirestimatedpopu-
lation grow th rate (λ: 1.01, SD: 0.19) was consistent with cou nt-based sur veys (λ:
0.986).Simulationsindicatedthatreliable(CV<0.10)estimatesoffecunditycouldbe
obtained with>1,000within-seasonliveencounters.Although precisionofsurvival
estimatesdependedprimarilyonnumbersofadultrecoveries,estimatesoffecundity
and population growth were most sensitive to total number of live encounters.
Synthesis and applications:Large-scalering-recoveryprogramscouldbeusedtoesti-
mate annual fecundity inmany species ofbirds, but the approach requires better
datacuration,includingaccurateassessmentofage,betterreportingofbandingto-
tals, and greater emphasis on obtaining and reporting within-season live
encounters.
KEY WORDS
Brownieband-recoverymodel,recruitment,Sebertag-recoverymodel
2 
|
   ARNOLD
1 | INTRODUCTION
Tag-recovery (a.k.a. ring- or band-recovery) models are widely
usedtoestimateannualsurvivalusingdataonnumbersofin divid-
uals surviving different intervalsbet weentag gingand reported
time of death (Brownie, Anderson, Burnham, & Robson, 1978;
Seber, 1970). Unlike live encounter data from restricted study
areas,whichprovide estimatesofapparentsurvivalφ=(1– mor-
ta li ty)*(1–p er manen te migrati on),de adrecover ydatacanprovid e
estimatesoftruesur vivalS =(1–mortality)providedthattagre-
coveriesoccurfromthroughoutthepopulation’spotentialdisper-
salormigr atoryra nge.Thisoccursmostcommonlywithharves ted
populationsofbirdsandfish(Brownieetal.,1978;Pollock,Hearn,
& Polacheck, 2002), although dead-recovery models have also
beenappliedto unhar vestedspecies(Francis,1995;Siriwardena,
Baillie, & Wilson,1998). Dead recoveries c an also be combined
with live-encounterdata fromrestrictedstudyareastoestimate
true survival andpermanentemigration(Barker,1997;Burnham,
1993).
Ifindividualscanbereliablyassignedtoageclassesatthetimeof
marking,tag-recovery modelscanbeusedto estimateage-specific
survival (Brownie etal., 1978; Pollock etal., 2002; Seber, 1971).
Most typically with birds, this approach has been used to provide
age-specificsurvivalestimatesforjuveniles(Sj)andadults(Sa),butit
canalsobeusedforthreeormoreageclassesprovidedageclasses
can be rec ognized at marki ng (Brownie etal. , 1978). For mon oga-
mousspecies that reachsexualmaturityasone-year-oldsandhave
limited se x- or age-specific variat ion in survi val or fecundit y (e.g.,
manysmall birds and mammals),populationdynamicscanbemod-
eledusing asimpleone-stageprojectionmodelthat capturesmost
oftheimportantvariationinvitalrates:
where Sa,t + FtSj,tisthepopulationgrowthrate,
𝜆t=Nt+1Nt
(Pulliam,
1988). Populations with additional nonbreeding stages could be
readily modeledbyextending this framework to include sub-adult
survivalandlifestages.Thus,tag-recoverymodelscanprovideeve-
rythingneededtoestimateλtexceptannualfecundityFt.
Fecundit yc anbeestimatedusingageratios(Nj ,t/Na,t)collected
duringpostbirthpulsesurveys,andageratiosarecommonlyused
when juveniles and adults can be readily distinguished during
survey counts (Harris, Kauffman,&Mills, 2007;Weegman etal.,
2016).Wildlife managers have long usedageratios of har vested
individuals (Hj,t/Ha,t) to measure annual fecundity, but because
juvenile s are often mor e vulnerab le to harves t than adult s, tag-
recoverydataareneededtoadjustthesedataforrelativevulner-
abilitytoharvest:
where
̂
fj,t
̂
f
a,t
 is the ratio of juvenile to adult harvest rates
(Zimmermanetal., 2010). Age ratiosat capture can providesimilar
estimates of population-level fecundity (Specht & Arnold, 2018),
but if capture proba bilities differ b etween age classes, fecundity
estimateswillbe biased.However,liverecapturesduring the initial
bandingperiodcouldbeusedtoassessage-specificvulnerabilityto
captureand estimatethetrue underlying agedistribution,similarly
to vulner ability-adjusted age rati os at harvest ( Alisauska s, Arnold,
Leafloo r,Ot is, & Sedinger, 2014; Zimme rman etal., 2010). Even if
estimationofcapturevulnerabilityisnotpossible,ageratiosatcap-
turemightneverthelessprovideareliableindexofannualfecundity.
Althoughuncorrectedageratiosatcapturehavebeenusedtoassess
population-levelfecundity(Mazerolle,Dufour,Hobson,&denHaan,
2005; Ross, Alisauskas,Douglas, &Kellett,2017;Specht& Arnold,
2018),models to estimate fecundity frominitial capture data have
notbeenformallydevelopedfortag-recover ystudies,althoughLink
and Barke r (2005) have addres sed this issue for op en-popul ation
mark–recapturedata.
Myobjectivesinthis studyaretodevelop and applypopula-
tion projection modelsincludingfecundity,juvenile survival, and
adult survivalderived solely from taggingdata (i.e., age-specific
counts of numbers of birdsbandedduring postbreeding capture
occasions, recaptured alive during the same season as originally
marked or subsequently found dead any time after marking).
I apply these models to two species of North American birds.
Northern pintails (Anas acuta; Figure1)have experienced apro-
longedpopulationdeclineandpreviousstudieshaveshownitcan-
notbeexplainedbydecliningsurvival(Bartzen & Dufour,2017),
sugges ting that lowered fe cundity may be th e cause (Specht &
Arnold,2018),buttodatetherehavebeennointegratedanalyses
forpintail stoidentifyrel ativecontribut ionsofdifferentvitalr ates
toobservedpopulationchanges(Koons,Arnold,&Schaub,2017).
Dark-eyedjuncos(Junco hyemalis)areawidespreadpasserinethat
hasbeenwell-studiedatseverallocalizedandprimarilysouthern
breeding sites (Nolan. etal., 20 02), but most of their breeding
rangeoccurs in remote portions oftheboreal forestthatliewell
nort h of establishe d monitoring prog rams (Saracco, De Sante, &
Kaschube,2008;Sauer&Link ,2011).Howeve r,theyarewel lsam-
pledbymi gr antbandin gstations(Leppold&Mulvihill ,2011),soan
Nt+1
=N
t(
S
a,t
+F
t
S
j,t)
̂
F
t=
H
j,t
H
a,t
̂
f
j,t
̂
f
a,t
FIGURE1 Afemalenorthernpintail(Anas acuta)withthree
nearlyfledgedducklings.Photographcredit:FredGreenslade,Delta
Waterfowl
    
|
 3
ARNOLD
approachthatcould estimate survival, fecundity,and population
trajectoryasbirdspasssouthwardduringfallmigration wouldbe
veryusefulforpopulationmonitoring,andcouldbeapplicableto
numerousotherHolarcticspecieswithextremenorthernbreeding
distrib utions (Huss ell & Ralph, 2 005; Spina , 1999).Mo del-ba sed
fecundityestimatesseemedreasonableforbothpintails andjun-
cos,butprecisionwaspoorgivensmallnumbersofwithin-season
recaptures,soIalsoconductedasimulationstudytoidentif ynec-
essarysamplesizesforobtainingmorepreciseestimates.Thisap-
proach provides new opportunities to estimate annualfecundity
atlocaltocontinentalscalesandcouldgreatlyleveragetheutility
of existin g banding data b y allowing invest igators to est imate a
completeensembleofvitalratesfromtaggingstudies.
2 | MODEL AND METHODS
Themodeldevelopedhereassumesthatanimalsarecapturedafter
thebreedingseasonhasendedusingmethodsthataresimilarlyef-
fectiveatcapturingadultsandyoungoftheyear,andthatcaptured
individualscanbereliablyagedattimeofmarking(e.g.,Pyle,1997).
2.1 | Model development
Anaïveestimatorofannual fecundity that ignoresdifferential vul-
nerabil ity to captur e is number of newly m arked juvenile fe males
divided by number of newly marked adult femalesMjF/MaF(or for
sexuallymonomorphicspecies,Mj/Ma).Theequationforestimating
ageratiosfromharvestdatacanbe modifiedfortag-recoverydata
as:
where
̂
pjF,t
isestimatedcaptureprobability for juvenile females in
year t,
̂
paF,t
isthe equivalent parameter foradultfemales,and
̂
NjF,t
and
areHorvitzandThompson(1952)estimatorsofpopulation
sizeattimeofcapture.Anyappropriateclosed-populationmark–re-
capturemodelcouldbeusedtoestimate captureprobabilities, but
giventhesparsenessofrecapturedatainmyexamples,IusedChao’s
(1989)est imator,whichconditionsonthenumb erofindividualsc ap-
tured1versus2times.Under thismodel, vulnerabilityadjustedfe-
cundity(
̂
F
)canbeestimatedas:
where
̂
F
isthe estimated age ratio,
̂
NjF
and
̂
NaF
areestimatedpopu-
lationsof juvenile andadultfemalesthatwereavailableforcapture,
MjF and MaFwerethe totalnumbers ofjuvenilesandadults captured
andmarked(i.e., MjF/MaF is a naïveestimateoffecundity,estimated
as
F
naive =
c
jF
1c
jF
,wherecjF istheprobabilitythatan initialcaptureofa
femalewillbejuvenile:
MjF
Bin
(
M
jF
+M
aF
,c
jF)
),andf1jF,f2jF,f1aF,and
f2aFwerethenumbersofjuvenileandadult femalescaptured1or2
times,respectively(i.e.,MjF = f1jF + f2jF,f2jF ~ Bin(MjF,
̂
pjF
),MaF = f1aF +
f2aF,f2aF ~ Bin(MaF,
̂
paF
)).Relativevulnerability(
̂
V
)tocapturecanthen
beestimatedas:
Es t imat ionofa ger a tio sfro mli v ere c apt u red atar equi resa sim -
ilarsetofassumptionsassimpleclosed-populationmark-recapture
models(i.e.ModelM0;Otis,Burnham,White,&Anderson,1978),
namely that: (a)the population is closed duringsampling,(b)an-
imals do notlose marks, (c) all marks are reportedon discover y,
(d)withinagegroups,allindividualshavethesameprobability of
capture,and(e)markinganimalsdoesnotaffecttheirsubsequent
catchability.These assumptions havebeen treatedindetail else-
where (Ot is etal., 1978), so I focus here o n potential viol ations
specif ic to their appli cation for es timating age rat ios. Tos atisfy
theclosureassumption,analystsneedtoselectappropriatemark-
ingperiods forassessingpostbirthpulse agestructure;choosing
intervalsafteryounghavebecomemobile,butbeforepostbreed-
ing dispe rsal or diffe rential migrat ion have altered lo cal age ra-
tios.Ifdataarecollectedduringmigration ,thenringingoperations
shoul dincludet heentiremigratio nperiodsothatc aptu redat aa re
not affec ted by differ ential migratio n of adults versu s juveniles
(Kelly&Finch, 2000). Markerlossisnegligibleforwithin-season
recapt ures, but iro nically many No rth Amer ican bande rs do not
report within-season live encountersbe causethe Bird Banding
Labor atory histor ically discou raged such repo rts (Buckl ey etal.,
1998). Homogeneity of capture probabilities among individuals
and absen ce of behaviora l response to c apture are ass umptions
that can b e accommodated und er more elaborate m odels (Otis
etal.,1978),buttheseassumptionsaredifficulttotestwithsparse
data(Chao,1989).
2.2 | Application
Examplesu se dinth isstudyin cl udenor thernpint ai lscapturedpri-
marilyusingbaittraps(Figure2)ontheirNorthAmericanbreeding
groundsduringJulythroughSeptember(Bartzen&Dufour,2017)
anddark-eyedjuncoscapturedprimarilyusingmistnetsthrough-
outNorthAmericaduringAugust–Octobermigration(Hussell&
Ralph, 2005). Iused datafrom64,201juvenile and 62,341adult
female northern pintails banded throughout the United States
andCanadaduring1970–1993andshotorfounddeadduringthe
hunting season(1Sep-31 Jan of year t+1)1970–1993toassess
performanceofthefecunditymodel.Inadditiontothe3,841and
2,377 dead recoveries obtainedfrom juvenilesandadultsduring
subsequ ent hunting season s, there were 90 and 4 4 live recap-
turesrecordedduringtheinitial bandingseason. Forpintails,an-
nualsurveydatawereavailablefromtheWaterfowlBreedingPair
and Habitat Sur vey (U.S. Fish and Wildlife Service 2017), which
indicatedaseverepopulationdeclineduringthistimeperiod.For
dark-eyed juncos, dat a included 248,939 and 107,998 juveniles
̂
F
t=
̂
NjF,t
̂
N
aF,t
=MjF,tM
aF,t
̂
pjF,t̂
paF,t
̂
F
=
̂
NjF
̂
N
aF
=MjF +(f1jF)2(2 ×f2jF )
MaF +(f1aF)2(2 ×f2aF
)
̂
V
=
̂
p
jF
̂
p
aF
=
M
jF
M
aF
̂
F
4 
|
   ARNOLD
andadultsbandedbetween1955and2013,butonly121and68
dead recoveries and 45 and15liveencounters duringthe initial
bandingseason.
Isummarizeddataonnumberofbandings,deadrecoveries,and
liveencountersduring theinitial trappingperiodintom-arraysfol-
lowing standard procedures for band-recovery analysis (Brownie
etal., 1978), except I inc luded an addit ional vector f or recaptur es
(liveencounters)duringtheinitialcaptureperiod.Forjuncos,Isum-
marized data in collapsed m-array format recognizing only years
since band ing (Kéry & Sch aub, 2012: 256) bec ause data were t oo
sparsetoconsiderannualvariationin survivalorencounterproba-
bilities.Summarizedm-arraysandadditional detailsaboutthedata,
analysis, and JAGS code are provided as Suppor ting Information
(DataS1).
Asaninitialtemplateforanalysis,IusedSeber’s(1971)model
for estimating survival (S) and reporting rates (r) from dead-
recover y data, as coded by Kér y and Schaub (2 012) for analy-
sisinWinBUGSandfurthermodifiedforanalysisin JAGS 3.3.0
(Plummer, 2012) usingthejagsUI package in R(Kellner, 2015). I
first considered models that treated all parameters as constant
through time(Sj,Sa,rj, ra, pj,pa,cj).Forpintails,which hadmore
extensivedata,I alsoconsideredmodelsthatincludedtemporal
and age-specif ic variation in s urvival, re covery, and initial c ap-
ture probabilities (Sj,t, Sa,t, rj ,t, ra,t, cj,t), but re capture dat a were
toosparsesoIt reatedc apturep robabil ities(pj,pa)asage-specific
consta nts in all models. T ime constant par ameters were given
vagu eunif ormpr ior sonth er ealsc a le (i.e. ,U nifor m[0 ,1] ), where as
temporallyvariableparametersweregiven vague priors on the
logit scale (mean ~ Uniform[−2,2], SD~Uniform[0,2]). For pin-
tails, I usedaninitial1000 iteration adaptation phase, followed
bythreeMarkovchainMonteCarlo(MCMC)chainsof25,000it-
erationseach ,wit ht hefi rst5,00 0iteratio nsdisc ar deda sb ur n- in,
andretai ni ngever y10t hi te rationforsamplingfr omth ep os terior
distrib ution. For junco s, I increased al l iterations by 10-fold to
accommodatesparsedata.Convergencewasachievedforallpa-
rameters (
̂
R
<1.01) with ru n times of <1min. Vulnerabi lity (V),
annualfecundity(Ft) andfinite populationgrowth(λt)were esti-
matedasderivedparameters:
2.3 | Simulation specifications
I conducted 1,000 24-year simulations representing a data-rich
scenario p atterned r oughly on the n orther n pintail dat a. For each
simulationIkeptSa,Sj and Fcons ta ntat0.60,0 .5 0,and0. 80 ,respec-
tively (hence,
λ=Sa+SjF=1
), but varie d number of recove ries and
recapturesusingrandomuniformdistributions on r(rj~U[0.0001,
0.4]; ra~U[0.00 01,0.2]),p(pa~U[0.0001,0.02])andV(U[0.5,1.5],
with pj = V × pa) to pro duce varyin g numbers of ban dings and live
anddeadencountersforfixedpopulationsizesofNj=240,000and
Na=300,000.Inadditiontoestimatesofthemeanandstandardde-
viation (SD),Icalculatedbias,coefficientofvariation(CV)androot
mean-squa red error (RMS E=√[bias2 + SD2]) for all po pulation and
encounterparameters.Icompared the accuracy (RMSE)andpreci-
sion(CV)oftheseestimatestonumbersofjuvenilesandadultsthat
werebanded,recapturedduringthefirst seasonfollowing banding
orrecovereddeadduring theirfirstorsubsequent years tocharac-
terizehowparameterestimateswereaffectedbyvariationinquan-
tityofdata.
3 | RESULTS
3.1 | Case studies
Juvenile pintails were more likely to be recaptured than adults
(
̂
V
=2.05,90%credibleinterval[CRI]:1.49–2.73),butuncertaintyin
thisparametertranslatedintolargeuncertainty inestimatesofad-
justedfecundityandpopulationgrowthrate.In thesimplestmodel
withnotemporalvariationinsurvivalandrecoveryrates,unadjusted
ageratios and λwerepreciselyestimated(CV<0.1);butrecapture
rates,vulnerability,andadjustedageratiosallhadCVsbetween0.1
and0.2(Table1).Ina fullytemporalmodel,adultsurvivalaveraged
0.601 with essentially no annual variation (SDt=0.002), juvenile
survivalaveraged0.654withmodestannualvariation(SDt = 0.064)
and fecundity averaged 0.520 with extensive annual variation
(SDt=0.227). Annual variation in λt was strongly correlated with
estimatedfecundity(Pearson’sr = 0.97), butnotwithadultorjuve-
nilesurvival(Figure3).Meanannual populationgrowth underboth
models (time constant: λ=0.93,90%CRI:0.841.04;timevarying:
V=̂
pĵ
pa
F
t=
̂
c
j,t
(1̂
c
j,t
)
V
λt=Sa,t +FtSj,t
FIGURE2 Ageratiosoffemalenorthernpintailscapturedfor
bandingcouldprovideestimatesofpopulation-levelfecundity,
butestimatesshouldbecorrectedforpotentialdif ferential
vulnerabilitytocapture.Photographcredit:DavidJohns
    
|
 5
ARNOLD
t=0.94,SDt=0.14)includedtheestimatederivedfromsurveydata
(
̂
λ
=0.945,SE = 0.001).
Forjuncos,juvenilevulnerabilitytocapturewas impreciselyes-
timated wi th a credible inte rval that overl apped 1 (
̂
V
=1.33, 90 %
CRI:0.79–2.10).Onlyunadjusted(raw)ageratiosandadultsurvival
werepreciselyestimated(CV<0.1),withremainingparametershav-
ingCVsexceeding 0.12(Table1).Estimatedλ was1.015(90%CRI:
0.755–1.371), which inc luded the continental estimates based on
Breedin g Bird Survey d ata (
̂
λ
0.989,95%CRI: 0.983–0.995;Sauer
&Link,2011)andconstant-effortringingstations(
̂
λ
1.0 07,95%CI:
0.997–1.017;DeSante,Kaschube,&Saracco,2015).
3.2 | Simulations
Precisionandaccuracy(i.e.,lowerCVandRMSE,respectively)ofju-
venile and adult survival and recovery probabilities increased with
increasing juvenile, adult, and total recoveries, but these relation-
ships were s trongest for a dult recoverie s (Figure S1). Accurac y of
vulnerabilityand adjustedfecunditywasmoststrongly affectedby
total number of live encounters (Figure4). Reasonable estimates
(CV<0.20)ofthesetwoparametersrequired>300totalrecaptures,
whereas preciseestimates (CV<0.10)required>1,000totalrecap-
tures. Because estimates of population growth (λ) depended on
bothsurvivalandfecundity,accuracy ofλestimateswereaffected
by both recove ries and rec aptures, but r ecaptures h ad a stronger
influence.
4 | DISCUSSION
U si n g e m p i ri c a l d a t a on a g e r a ti o s a t c ap t u r e f o rn o r t h e r n p in t a i l s a nd 
dark-e yedjuncos ,Iwasab letoo bt ai nest imatesofannualfe cundit y
Northern pintails Dark- eyed juncos
Mean SD CV Mean SD CV
S.juv 0.629 0.021 0.033 0. 276 0.055 0.200
S.ad 0. 613 0.0054 0.009 0.493 0.034 0.070
r.juv 0.097 0.006 0.061 0.00045 0.00006 0.138
r.ad 0.0412 0.0008 0.020 0.0 0063 0.00008 0.122
p.juv 0.00144 0.00015 0.105 0.00019 0.00003 0.147
p.ad 0.00072 0.00011 0.153 0.00015 0.00004 0.250
V2.04 0.38 0.187 1.330 0.410 0.308
Mj/Ma0.992 0.0057 0.006 2.305 0.008 0.004
F0.503 0.094 0.186 1.889 0.55 6 0.294
λ0.929 0.060 0.064 1. 014 0.191 0.189
TABLE1 Estimatesofjuvenile( juv)
andadult(ad)annualsurvival(S),dead
recovery(r)andliverecapture(p)
probabilitiesandassociatedestimatesof
capturevulnerability(V),ageratiosat
capture(Mj/Ma),fecundity(F)andfinite
populationgrowth(λ)fornor thernpintails
anddark-eyedjuncosundertime-constant
models
FIGURE3 Annualestimates(90%credibleintervals)of
juvenilesurvival(JuvSurv;green,closedsymbols),adultsur vival
(AdSur v;blue-green,opensymbols),fecundity(purple)andannual
populationgrowth(Lambda;black)fornorthernpintailsduring
1970–1993.Fecunditywasestimatedfromageratiosatcapture
andexplainedmostoftheannualvariationinlambda
FIGURE4 Effectofnumberofliverecaptures(combined
juvenileandadult)duringtheinitialmarkingperiodonrootmean-
squarederror(RMSE)ofvulnerabilitytocapture(Vuln),annual
fecundity(F.adj)andpopulationgrowthrate(lambda).Notethat
agecompositionofrecaptures(0.5–1.5juvenilesperadult)and
deadrecoveryprobability(0–0.4forjuveniles,0–0.2foradults)
alsovariedrandomlyamongsimulations,addingtoamong-replicate
variation
6 
|
   ARNOLD
byadjustingforvulnerability to captureusing live encounters ob-
tained d uring the original banding se ason. These fecu ndity esti-
matescomplemented estimates ofjuvenile andadultsurvivalthat
analysts have typicallyobtainedfrom tag-recovery data(Brownie
etal., 1978;Siriwardenaetal.,1998)andallowed me toconstruct
modelst hatincludedallofthedemographiccomponentsofpopula-
tion grow th. For pinta ils, data were s ufficient t o estimate ann ual
variationinallthreevitalratesandtheseestimatessuggestedthat
obser ved populati on declines du ring 1970–1993 were drive n pri-
marily by reductionsinannual fecundity, which is consistent with
other rece nt studies of his torical pinta il data (Bart zen & Dufour,
2017;Specht&Arnold,2018).Forjuncos,myestimateofadultsur-
vival(0.493,SD:0.034)wasslightlylargerthanestimatesofappar-
entsurvivalfromconstant-effortringingstations(0.453,SD: 0.030;
DeSanteetal.,2015),whereasmyestimateofrecruitment(F × Sjuv
=1.889×0.276=0.521)wasslightlylowerthanestimatesderived
fromreverse-timeanalyses(0.554,SD:0.032;DeSanteetal.,2015).
Bothoftheseslightdifferencesare consistentwiththerealization
that in Pra del (1996)mo dels, appare nt surviva l estimates in clude
permanent emigration and mortality, whereas recruitment esti-
matesincludebothimmigrationandfecundity.
Avianecologists oftenhave accesstolarge-scalecountdata to
assessannualfluctuationsinpopulationsize(Newson,Evans,Noble,
Greenwood, & Gaston, 2008; Sauer&Link, 2011)andcontinental
bandingorringingprogramscanprovidesimilardataonage-specific
survivalorapparentsurvival(Francis,1995;Saracco,Royle,DeSante,
& Gardner, 2010; Sir iwardena eta l., 1998), but fecundi ty data are
oftenlacking(Ahrestani,Saracco,Sauer,Pardieck,&Royle,2017).To
assess fecundity, population modelers have usedageratios at har-
vest from hunted species(Péron & Koons, 2012), fledgling counts
from citizen-scientist nest-record programs (Robinson, Morrison,
& Baillie, 2 014), data f rom small-scale ne sting studi es (Weegman,
Arnold,Dawson,Winkler,&Clark,2017)andreverse-timemark–re-
capturemodels(whichmeasure theproduct offecundityandfirst-
year survival; Pradel, 1996;Saracco etal., 2008), but age ratiosat
capturecouldprovideanalternativeorcomplementarydatastream
to assess spatiotemporal variation in fecundity (Mazerolle etal.,
2005; Ross etal.,2017;Specht&Arnold,2018).Inthe absenceof
liverecapturedata, vulnerabilitytocapture(V) couldbeestimated
in an integr ated population m odeling (IPM) fra mework (Ahre stani
etal., 2017), assuming that auxiliary population count data were
availableandthattherewerenoconfoundinginfluencesofimmigra-
tionoremigration:
If markin g efforts oc cur at the end of the bre eding season,
butbeforepostbreedingdispersalormigration,thenageratiosat
marking measure local reproductivesuccess and spatially exten-
sive marki ng data have the pot ential to measure r egional varia-
tioninfecundityandidentifyecologicaloranthropogenicdrivers
of this vari ation (Specht & A rnold, 2018). However, rese archers
must haveathoroughunderstanding of breedingand movement
phenology to select appropriate intervals and spatialscales for
dataanalysis,therebyassuringthatageratiosarenotaffectedby
on go ingbr eed ingef for t sorea rly dispe rsalormi gra tionbyo neage
class versusanother (Andres, Browne, &Brann,2005). Variation
inageratiosatcapturecouldalsobeduetoage-relatedvariation
inlocalhabitatuseonthebreedinggrounds,especiallyifcapture
efforts are not randomly distributed among potential habitats.
Treating individual capture sitesasrandomeffects couldpoten-
tiallycontrolforsomeofthislocation-specificvariation(Specht&
Arnold,2018)andtesting for seasonal trendsinageratioscould
helpidentifyongoingbreedingordifferentialmovements.Agera-
tios might a lso be affec ted by capture m ethods, if ju veniles are
more(orless)vulnerable tocaptureby widelyemployedcapture
methods. In North America ,relatively few capture metho dsare
uniquely coded at time of banding (https://www.pwrc.usgs.gov/
BBL/MANUAL/summary.cfm),butEuropeanringingschemesre-
cordawidediversityofcapturemethodsandluretypes(https://
euring.org/files/documents/E2000PLUSExchangeCodev1161.
pdf),therebyallowingforathoroughinvestigationofheterogene-
ityinageratiosinducedbycapturemethodology.
In the nor thern hemis phere, many birds a re banded or ring ed
duringautumnastheymigratefromnorthernhemispherebreeding
sitestoequatorialorsouthernhemispherewinteringsites(Hussell&
Ralph, 2005;Spina,1999).Suchmarkingprogramshavethepoten-
tialtoassesscontinental-levelproductivity,butmeetingtheclosure
assumptionseemsmuchmoredifficultinthissituation(Hochachka
&Fiedler,2008).Nichols,Thomas,andConn(2009)partitionedde-
tectio n probabilit y from count sur veys into four condit ional com-
ponents,andasimilarhierarchycouldbeextendedtosame-season
captureprobabilities.First,choiceofmarkingsitescouldaffectage
ratios at capture if juveniles and adults have different migration
routes (Ralph,1978).Second,differential timingof migration could
affec t age ratios at c apture (An dres etal., 20 05), especi ally if one
ageclass exhibits amore prolonged migration andcaptureefforts
arelimitedtoperiodsofpeakmigration.Third,age-relatedcapture
probability could be affected by differencesinstopover durations;
forexample, ifjuvenilesspendmoretime“refueling”atmigrational
stopover si tes they would be m ore vulnera ble to capture ( Rguibi-
Idrissi, J ulliard, & Bair lein, 2003), e specially if pe rmanent mark ing
sitesareconcentratedatmigrationalstopoversites.Finally,because
juvenilesaremorenaïve,theymaybemorevulnerabletocaptureby
standardtrappingmethods(Rguibi-Idrissietal.,2003),evenifloca-
tionsandtimingwereotherwiseunbiased.
Probably the biggest limitation to employing the fecundity
estimat ion approach deve loped herein is t he paucity of wit hin-
season live-encounter data for estimating vulnerability to initial
captur e. With suff icient reca pture data, m any of these assu mp-
tions coul d be tested, an d some ringin g stations have s ufficient
in-housedatatoestimatecapturevulnerability(e.g.,Hochachka&
Fiedler,200 8).D uringroutineduckbandingoperationsinAlberta,
Canada,approximately54%of33,552duckscapturedforbanding
overa three-year period were within-seasonrecaptures (Dieter,
Murano, & Galster, 2009), but banding crews have not been
Nt+1=Nt[Sa+(SjMj)(VMa)]t
    
|
 7
ARNOLD
encouraged to collect and report these data. North American
banderswerehistorically dissuadedfromreportingsame-station
live encounters,and hence, live encounter data arelimiting for
historical analyses, althoughthis shortcoming has been recently
corrected(Smith, 2013) and many NorthAmericanbandershave
begun sub mitting large amo unts of recapture data (D. Bystrak,
PatuxentWildlifeResearchCenter,pers.comm.).InEurope,many
national r inging program s failed to keep reco rds of numbers of
bands deployed and focused primarily on banding known-age
juveniles,butthis shortcoming was recognizedinthe mid-1980s
(e.g., And erson, Burnha m, & White, 1985) and ringing schemes
havesince expanded to include adults, and historical summaries
ofringdeploymenthavesincebeencompiledformany European
countriesgoingbackto1975(https://euring.org/data-and-codes/
ringing-totals).Birdringersneedtobemadeawareofthevalueof
liveencounters, even thosefrom the samelocationand banding
season , and national ban ding program s need to be made awa re
ofthe valueofcollectingand archiving suchdata. Theability to
estimatefecundity fromageratiosatthe time ofmarkinggreatly
enhance s the utility of c ontinental r inging progra ms, because i t
allows imp ortant vit al rates to be es timated as marke rs are de-
ployed,whileinvestigatorswaitforencounterdatatoaccumulate.
ACKNOWLEDGMENTS
I am gratefu l to the countle ss individua ls who deploye d bands on
pintailsandjuncos,andtoallthehuntersandcitizenswhoreported
encounters to the Bird Banding L aboratory. Hannah Specht, Ray
Alisauskas and an anonymous reviewer provided helpful feedback
onearlierdraftsofthemanuscript.
AUTHOR CONTRIBUTIONS
TWA conceived the idea, collated and analyzed data, developed
code,andwrotethemanuscript.
DATA ACCESSIBLITY
Data and code used for analysis and simulations are inclu ded as
SupportingInformation(DataS1).
ORCID
Todd W. Arnold http://orcid.org/0000-0002-7920-772X
REFERENCES
Ahres tani, F. S., Sara cco, J. F.,S auer, J. R., Pard ieck, K. L ., & Royle, J.
A. (2017). A n integrated popu lation model for bird monitoring in
North America. Ecological Applications, 27, 916–924. https://doi.
org/10.1002/eap.1493
Alisauskas,R.T.,Arnold,T.W.,Leafloor,J.O.,Otis,D.L.,&Sedinger,J.S.
(2014).Lincolnestimatesofmallard(Anas platyrhynchos)abundance
in North America. Ecolog y and Evolution, 4, 132–143. https://doi.
org /10.1002/ece3.90 6
Ander son, D. R., Burnh am, K. P., & White, G. C. (1985). Prob lems in
estima ting age-specif ic survival r ates from recover y data of birds
ringed as yo ung. Journal of Animal Ecology, 54, 89–98. https://doi.
org/10.2307/4622
Andres, B. A., Browne, B. T., & Brann, D. L. (2005). Composition,
abundance, and timing of post-breeding migrating landbirds
at Yakutat, Alaska. Wilson Bulletin, 117, 270 –279. ht t ps: //do i .
org /10.1676/04-0 39.1
Barker, R. J. (1997). Joint modeling of live-recapture, tag-resight,
and tag-recovery data. Biometrics, 53, 666–677. https://doi.
org/10.2307/2533966
Bartzen,B.A.,&Dufour,K.W.(2017).Northernpintail(Anas acuta)sur-
vival, recovery,andharvestratesderived from 55years ofbanding
inPrairieCanada.Avian Conser vation and Ecology,12,7.ht tps://doi.
org/10.5751/ACE-010 48-120207
Brownie , C., Ander son, D. R., B urnham, K . P., & Robso n, D. S. (1978).
Statistic al inference from ba nd recovery data: A han dbook.Washington,
DC:U.S.FishandWildlifeServiceResourcePublication131.
Buckley,P. A., Francis,C. M.,Blancher,P.,Robbins, C. S., Smith, G., &
Cannell,P.(1998).The Nor th American Bird BandingProgram:Into
the21stcentur y.Journal of Field Ornithology,69,511–529.
Burnham,K.P.(1993).A theoryforcombinedanalysis of ring recovery
andrecapturedata. In J. D.Lebreton(Ed.),Marked individuals in the
study of bird populations(pp.199–214).Basel ,Switzer la nd :B ir kh aüser.
Chao, A.(1989).Estimatingpopulation size for sparse data in capture-
recapture experiments. Biometrics, 45, 427–438. https://doi.
org/10. 2307/2531487
DeSant e, D. F., Kas chube, D. R. &. , & Saracco, J. F. (2015). Vital rates
of North American Landbirds. The Institute for Bird Populations.
Retrievedfromwww.VitalRatesOfNorthAmericanLandbirds.org
Dieter,C .D., Murano,R .J.,&Galster,D.(20 09).Captureand mor talit y
ratesofducksinselectedtraptypes.Journal of Wildlife Management,
73,1223–1228.https://doi.org/10.2193/2008- 438
Francis,C.M.(1995).HowusefularerecoveriesofNorthAmericanpas-
serinesforsurvivalanalyses?Journal of Applied Statistics,22,1075–
1081.https://doi.org/10.10 80/02664769524838
Harris , N. C., Kauf fman, M. J., & Mi lls, L. S. (20 07). Inference s about
ungulate p opulation dynamics derived from age ratios. Journal of
Wildlife Management,72,1143–1151.
Hochachka, W. M., & Fiedler, W. (2008). Trends in trappabilit y
and stop-over duration can confound interpretations of pop-
ulation trajectories from long-term migration ringing studies.
Journal of Ornithology, 149, 375–391. ht tps://doi.org/10.1007/
s10336-008-0282-1
Horvitz, D. G., & Thompson, D. J.(1952).A generalization of sampling
withoutreplacementfroma finiteuniverse.Journal of the American
Statistical Association,47,663–685.https://doi.org/10.1080/016214
59.1952.1048 34 46
Hussell,D.J.T.,&Ralph,C.J.(2005).Recommendedmethodsformoni-
toringchangeinlandbirdpopulationsbycountingandcapturingmi-
grants.North American Bird Bander,30,6–20.
Kellner,K.(2015).jagsUI: a wrapper around rjags to streamline JAGS analy-
ses.Rpackageversion1.3.7.Retrievedfromhttp://CRAN.R-project.
org/package=jagsUI
Kelly,J.F.,&Finch,D.M.(2000).Effectsofsamplingdesignonageratios
ofmigrantscapturedatstopoversites.Condor,102,699–70 2.h ttps ://
doi.org/10.1650/0010-5422(2000)102[0699:EOSDOA]2.0.CO;2
Kéry, M., & Schaub, M. (2012). Bayesian population analysis using
WinBUGS, A hierarchical perspective(1sted.).Elsevier,MA:Academic
Press.
Koons, D. N. , Arnold, T. W., & Schaub, M . (2017). Unders tanding the
demographic driversofrealized population growth rates.Ecological
Applications,27,2102–2115.https://doi.or g/10.1002/eap.1594
8 
|
   ARNOLD
Leppold,A .J.,& Mulvihill, R.S. (2011). The boreallandbird component
ofmigrantbirdcommunitiesineasternNorthAmerica.InJ.V.Wells
(Ed.), Boreal birds of North America: A hemispheric view of their con-
servation links and significance(pp. 73–84). Berkeley, CA: Studies in
AvianBiology,41,UniversityofCaliforniaPress.
Link, W. A., & Barker, R. J. (2005). Modeling association among
demographic parameters in analysis of open population
capture-recapture data. Biometrics, 61, 46–54. https://doi.
org/10.1111/j.0006-341X.2005.030906.x
Mazerolle,D.F.,Dufour,K.W.,Hobson,K.A.,&denHaan,H.E.(2005).
Effects of large-scale climaticfluctuations onsurvival and produc-
tionofyounginaNeotropicalmigrantsongbird,the yellow warbler
Dendroica petechia. Journal of Avian Biology,36,155–163.https://doi.
org/10.1111/j.0908-8857.2005.03289.x
Newson,S.E.,Evans,K.L., Noble,D.G.,Greenwood,J.J. D.,&Gaston,
K. J. (20 08). Useofdistance sampling to improve estimates of na-
tional populationsizes for commonand widespread breeding birds
in the UK . Journal of Applied Ecology, 45, 1330–1338. https ://doi .
org/10.1111/j.1365-2664.2008.01480.x
Nichols,J.D.,Thomas,L.,&Conn,P.B.(2009).Inferencesaboutlandbird
abundancefromcountdata:Recentadvancesandfuturedirections.
Environmental and Ecological Statistics,3,201–235.
Nolan.,V.Jr,Ketterson,E.D.,Cristol, D. A., Rogers,C.M.,Clotfelter,E.
D.,Titus,R. C.,… Snajdr,E. (2002). Dark-eyed junco (Juncohyema-
lis), version 2.0 in Birds of North America.Ithaca, NY:CornellLab of
Ornithology.
Otis, D. L., Burnham, K. P., White, G. C., & Anderson, D. R. (1978).
Statisticalinferencefromcapturedataonclosedanimalpopulations.
Wildlife Monographs,62,1–35.
Péron, G.,& Koons, D.N.(2012). Integrated modeling ofcommunities:
Parasitism, competition, anddem ographicsynchrony in sympatric
ducks.Ecology,93,2456–2464.https://doi.org/10.1890/11-1881.1
Plumme r,M. (2 012). JAGS version 3.3.0 user manual. [online]. Retrieve d
http://mcmcjags.sourceforge.net/.
P ol l o c k , K .H . , H e a r n ,W . S ., & P o l ac h e c k , T. ( 20 0 2 ). A g e ne r a l m o d e lf o r t a g-
gingonmultiplecomponentfisheries:Anintegrationofage-dependent
reportingratesandmortalityestimation.Environmental and Ecological
Statistics,9,57–69.https://doi.org/10.1023/A:1013715008683
Pradel,R.(1996).Utilizationofcapture-mark-recapture forthestudyof
recrui tment and population grow th rate. Biometrics, 52, 703–709.
https://doi.org/10.2307/2532908
Pulliam,H.R.(1988).Sources,sinks,andpopulationregulation.American
Naturalist,132,652–661.https://doi.org/10.1086/284880
Pyle, P. (1997). Identification guide to North American birds, Part I:
Columbidae to Ploceidae.Bolinas,CA:SlateCreekPress.
Ralph, C . J. (1978). The disori entation a nd possib le fate of young p as-
serine coastal migrants. Bird- Banding, 49, 237–247. https://doi.
org/10.2307/4512365
Rguibi-Idrissi,H., Julliard,R .,&Bairlein,F.(2003).Variationinthestop-
overdurationof reedwarblersAcrocephalus scirpaceus in Morocco:
Effects of season, age and site. Ibis, 145, 650–656. ht tps://doi.
org/10.1046/j.1474-919X.2003.00208.x
Robinson,R. A.,Morrison, C. A ., & Baillie, S. R. (2014).Integratingde-
mographicdata: Towards a framework for monitoringwildlife pop-
ulationsat large spatial scales. Methods in Ecology and Evolution, 5,
1361–1372.https ://doi.org/10.1111/2041-210X .12 20 4
Ross, M. V., Alis auskas, R. T., Dou glas, D. C., & Kelle tt, D. K. (2017).
Decadal declines in avian herbivore reproduction: Density-
dependent nutrition and phenological mismatch in the Arctic.
Ecology,98,1869–1883.https://doi.org/10.1002/ecy.1856
Saracco,J.F.,DeSante,D. F.,&Kaschube,D.R.(2008).Assessingland-
bird monitoring programs and demographic causes of population
trends.Journal of Wildlife Management,72, 1665–1673. https://doi.
org/10.2193/20 08-129
Saracco,J.F.,Royle,J.A.,DeSante,D.F.,&Gardner,B.(2010).Modeling
spatialvariationinaviansurvivalandresidencyprobabilities.Ecology,
91,1885–1891.https://doi.org/10.1890/09-0705.1
Sauer,J.R.,&Link,W.A.(2011).AnalysisoftheNorthAmericanBreeding
BirdSurveyusinghierarchical models.Auk,128, 87–98.https://doi.
org/10.1525/auk. 2010.09220
Seber, G. A . F. (1970). Estim ating time-spe cific surv ival and repor ting
rates for adult birds from band returns. Biometrika, 57, 313–318.
https://doi.org/10.1093/biomet/57.2.313
Seber, G. A.F.(1971).Estimating age-specific survivalrates frombird-
band ret urns when the rep orting rate is co nstant. Biometrika, 58,
491–497.https://doi.org/10.1093/biomet/58.3.491
Siriwardena, G. M., Baillie,S.R.,& Wilson,J.D. (1998).Variation inthe
survivalrates ofsome Britishpasserineswith respect totheir pop-
ulation trends on farmland. Bird Study, 45, 276–292. https://doi.
org /10.108 0/000636598094 61099
Smith, G . J. (2013). The U.S . Geological Survey Bird Banding Laboratory:
An integrated scientific program supporting research and conser vation
of North A merican birds: U.S. Geolog ical Survey Open-File Repor t 2013–
1238,Retrievedfromhttp://pubs.usgs.gov/of/2013/1238/
Spe cht,H .M.,&Arnold,T.W.(2018).Bandi ngager atiosrevealthatprai-
riewaterfowlfecundityisaffectedbyclimate, densitydependence
andpredator-preydynamics.Journal of Applied Ecology,32, inpress.
https://doi.org/10.1111/1365-2664.13186
Spina, F. (1999). Value of ringi ng information for b ird conserva tion in
Europe.Ringing and Migration,19,S29–S40.https://doi.org/10.1080
/03078698.1999.9674209
U.S.FishandWildlife Service(2017).Waterfowl population status, 2017.
Washington,DC:U.S.DepartmentoftheInterior.
Weegman, M.D.,Arnold,T.W.,Dawson,R.D.,Winkler,D.W.,& Clark,
R. G. (2017). Integrated population models reveal local weather
conditi ons are the key dr ivers of popu lation dyna mics in an aer ial
insectivore. Oecologia, 185, 119–130. https://doi.org/10.1007/
s00442-017-3890-8
Weegman, M . D., Bearhop, S ., Fox, A. D., Hilt on, G. M., Walsh , A. J.,
McDona ld,J.L .,&Hodgs on,D.J.(2016).I nte gratedpopul ati onm od-
ellingrevealsaperceivedsourcetobeacrypticsink.Journal of Animal
Ecology,85,467–475.https://doi.org/10.1111/1365-2656.12481
Zimmerman,G.S.,Link,W.A.,Conroy,M.J.,Sauer,J.R.,Richkus,K .D.,
&Boomer,G. S. (2010). Estimating migratorygame-bird productiv-
ity by integratingage ratioand banding data.Wildlife Research,37,
612–622.https://doi.org/10.1071/WR10062
SUPPORTING INFORMATION
Additional suppor ting information may be found online in the
SupportingInformationsectionattheendofthearticle.
How to cite this article:ArnoldT W.Usingring-recoveryand
within-seasonrecapturedatatoestimatefecundityand
populationgrowth.Ecol Evol. 20 18; 0 0:1–8.
https://doi.org/10.1002/ece3.4506

Supplementary resources (2)

... The proportion of juvenile females in the banding data does not necessarily represent the proportion of juvenile females in the population due to potential differences in the capture probability of juvenile and adult females. To correct for this potential bias, we created an additional parameter, denoted as q [obs] i,t , and let q [obs] i,t = i,t ∕ 1 + i,t , in which ν was the vulnerability parameter calculated as the ratio of the capture probability of juvenile females r [jf] to the capture probability of adult females r [af] (Arnold, 2018). The capture probability of juvenile females was estimated as r [jf] , in which B [jf] t was the number of banded juvenile females, and R [jf] t was the number of live reencounters in the same year of banding, all ecostrata combined. ...
... Because previous studies did not find evidence for age-specific survival (Arnold, 2018;Bartzen & Dufour, 2017), but see Ref. Flint, Grang, and Rockwell (1998, we only considered sex-specific survival and assumed that juveniles had the same survival probabilities as adults of the same sex. ...
... Nonetheless, band-recovery data have been used to only inform survival, despite the fact that the numbers of banded juveniles and adults are counts of different ages and thus can provide information for productivity estimation. In this study, we utilized banding information to estimate both productivity and survival rates, as demonstrated in recent studies (Arnold, 2018;Specht & Arnold, 2018). Even though such practices violate the assumption of independent data inputted to the IPM, there is evidence that dependent data will not lead to increased bias or uncertainty in parameter estimates (Abadi, Gimenez, Arlettaz, & Schaub, 2010 ...
Article
Full-text available
1.Knowledge of land use patterns that could affect animal population resiliency or vulnerability to environmental threats such as climate change is essential, yet the interactive effects of land use and climate on demography across space and time can be difficult to study. This is particularly true for migratory species, which rely on different landscapes throughout the year. 2.Unlike most North American migratory waterfowl, populations of northern pintails (Anas acuta; hereafter pintails) have not recovered since the 1980s despite extended periods of abundant flooded wetlands (i.e. ponds). The mechanisms and drivers involved in this discrepancy remain poorly understood. While pintails are similar to other ducks in their dependence on ponds throughout their annual cycle, their extensive use of croplands for nesting differentiates them and makes them particularly vulnerable to changes in agricultural land use on prairie breeding grounds. 3.Our intent was to quantify how changes in land use and ponds on breeding grounds have influenced pintail population dynamics by developing an integrated population model to analyse over five decades (1961–2014) of band‐recovery, breeding population survey, land use and pond count data. We focused especially on the interactive effects of pond counts and land use on pintail productivity, while accounting for density dependent processes. 4.Pintail populations responded more strongly to annual variation in productivity than survival. Productivity was positively correlated with pond count and negatively correlated with agricultural intensification. Further, a positive interaction between pond count and agricultural intensification was insufficient to overcome the strong negative effect of agricultural intensification on pintail productivity across nearly all pond counts. The interaction also indicated that pintail populations were more negatively impacted by the decrease in ponds associated with climate change under higher agricultural intensification. 5.Our results indicate that pintail populations have become more vulnerable to climate change under intensified land use, which suggests that future conservation strategies must adapt to these altered relationships. The interactive effects of land use and climate on demography should be considered more frequently in animal ecology, and integrated population models provide an adaptable framework to understand vital rates and their drivers simultaneously. This article is protected by copyright. All rights reserved.
... was the vulnerability parameter calculated as the ratio of the capture probabilities of juvenile females r [jf] to adult females r [af] (Arnold, 2018; Appendix 1). We only considered constant r [jf] and r [af] and used informative priors on these parameters (details see Appendix 1) due to limited live-encounter data. ...
... Because there is little evidence that adult waterfowl survival varies by upland cover types (Howerter et al., 2014), we did not include upland habitats as covariates in the survival sub-model. Also, based on prior analyses that found little evidence of age effects in survival (Arnold, 2018;Bartzen and Dufour, 2017), we assumed that juveniles had the same survival probabilities as adults of the same sex. Survival was estimated using a Brownie band-recovery model (Brownie et al., 1985). ...
Article
Accounting for the spatial variation of environmental drivers and demographic mechanisms in population predictions is essential for conservation prioritization under climate and land use changes but is often ignored. We developed an integrated population model (IPM) with region-specific population processes and used the model to prioritize region-specific conservation strategies for northern pintail ( Anas acuta ; hereafter pintail). Pintail are of high conservation concern in North America due to low productivity related to extensive use of cropland for nesting and wetland (pond) loss related to anthropogenic disturbance and climatic variability. We analyzed 25 years (1990–2014) of pintail breeding population survey, band-recovery, pond count, climate and land use data to estimate regional demography- environment relationships. We then predicted regional population responses under potential future changes in climate, wetland drainage, and agricultural intensification. Our IPM predicted that pintail populations will be sensitive to climate changes throughout the entire study area. Drainage was predicted to have more deleterious impacts in Parkland regions due to more extensive wetland drainage in these regions. Agricultural intensification was predicted to have more deleterious impacts in Saskatchewan-Prairie due to a stronger response of pintail productivity to agricultural intensification in this region. Our study highlights the importance of considering region-specific conservation strategies to accommodate regional variation in future global changes and demographic response to such changes. Our IPM that accommodates spatial variation in environmental changes and demographic responses is flexible for other systems, and thus is highly relevant to diverse studies in conservation prioritization given global change.
... Even with survey improvements, monitoring segments of the midcontinent population of sandhill cranes would require different methods because subspecies and breeding segments cannot be separately monitored with the current spring survey. Initiation of a banding program could be one source of data to meet these needs and provide considerable additional information useful for management (Alisauskas et al. 2013, Arnold 2018. Reliable subspecies determination can be made using morphological measurements of live birds during banding (VonBank et al. 2019), allowing for estimation of subspecies-specific survival and harvest rates. ...
Article
Sandhill cranes (Antigone canadensis) inhabiting the midcontinent of North America have been hunted since the 1960s under management goals of maintaining abundance, retaining geographic distribution, and maximizing sustainable harvest. Some biologists have raised concerns regarding harvest sustainability because sandhill cranes have lower reproductive rates than other game birds. We summarized demographic information in an age‐structured matrix model to better understand population dynamics and harvest. Population indices and recovered harvest since the early 1980s suggest midcontinent sandhill cranes have experienced an average long‐term annual growth of 0.9%; meanwhile, harvest has increased 1.8% annually. Adult survival and recruitment rates estimated from field data required modest adjustments (1–3%) so that model‐derived growth rates matched growth estimated from a long‐term survey (0.887 adult survival and 0.199 females/breeding female). Considering 0.9% long‐term annual growth, sandhill cranes could be harvested at a rate of 6.6% if harvest was additive to natural mortality (assumed to be 0.05) or 11.3% if harvest mortality compensated for natural mortality. Life‐history characteristics for long‐lived organisms and demographic evidence suggested that hunter harvest was primarily additive. Differential harvest rates of segments of sandhill cranes in the midcontinent population derived from differential exposure to hunting suggested potentially unsustainable harvest for greater sandhill cranes (A. c. tabida) from 2 breeding segments. Overall, demographic evidence suggests that the harvest of sandhill cranes in the midcontinent population has been managed sustainably. Monitoring activities that reduce nuisance variation and estimate vital and harvest rates by subspecies would support continued management of sandhill cranes that are of interest to hunters and bird watchers. Published 2020. This article is a U.S. Government work and is in the public domain in the USA. Long‐term survey estimates and population model results suggest that harvest of sandhill cranes from the midcontinent population has been implemented sustainably. Managing this population as a single entity involves potential risks to specific groups of sandhill cranes; therefore, improved monitoring activities that reduce nuisance variation and estimate vital and harvest rates would support continued management.
Article
Full-text available
Despite declines in numerous migratory bird populations due to global climate and landscape changes, the Pacific Flyway population of Greater White‐fronted Geese Anser albifrons frontalis in North America has flourished over recent decades. However, the demographic foundations of the population increase remain unclear, largely due to sparse data. In this study, we used a Bayesian integrated population model (IPM) to maximize information from multiple data sources including coordinated population survey, ring‐recovery and hunter harvested goose tail data. We estimated demographic parameters and assessed the role of several possible drivers of the observed population increase, including density dependent processes, agricultural land use change, and climate conditions in both wintering and breeding season, while also accounting for the impacts of harvest.
Article
Full-text available
Substantial effort has been dedicated to developing reliable monitoring schemes for North American bird populations, but our ability to monitor bird populations in the boreal forest remains limited because of the sparsity of long-term data sets, particularly in northerly regions. Given the importance of the boreal forest for many migratory birds, we set out to (1) summarize the main challenges associated with monitoring avian populations, (2) describe the available statistical tools for population monitoring and their applications, and (3) identify future directions to overcome current challenges in monitoring bird populations in the boreal forest. Defining and delineating populations of interest and identifying the drivers that affect those populations present the greatest current challenges. This is because migratory birds may be affected by many population-limiting processes at different stages of their annual life cycles. These factors are often hierarchically structured and can influence populations at the local, regional, or continental scales. Some of the challenges associated with delineating populations and identifying population drivers can be addressed via the plethora of sampling and analytic methods available to examine population change over time. Choosing the proper analytic methods depends on the goals of the study and the nature of the data such as single or multiple populations, repeated occurrence or count-based surveys, or demographic rates. Recent advances in hierarchical and integrated population models make these analytic approaches some of the most promising avenues for the development of future methods. However, these tools require large data sets, and acquiring sufficient data on bird populations and potential explanatory variables is difficult in the boreal forest. If the current challenges to monitoring birds in the boreal forest are to be overcome, serious effort should be dedicated to integrating existing data and making them accessible. Enhancing survey effort through multispecies surveys will also play an important role. Implementing spatially balanced sampling plans with a rotating panel design could balance the trade-offs between spatial versus temporal replication at an affordable cost. Improving the accessibility of environmental covariates that are spatially and temporally explicit would also enable development of mechanistic population models that improve our understanding of migratory bird population dynamics. Finally, given that long-term monitoring programs can take many decades before delivering reliable population trends and that organizational priorities often change over time, we suggest that collaborative efforts will help ensure the long-term survival of new monitoring programs. Surveillance des populations aviaires boréales : comment estimer les tendances et les trajectoires à partir de données bruyantes? RÉSUMÉ. Des efforts considérables ont été consacrés au développement de programmes de surveillance fiable des populations d'oiseaux d'Amérique du Nord. Toutefois, notre capacité à surveiller les populations aviaires dans la forêt boréale reste limitée, en raison de la rareté des jeux de données de longue durée, en particulier dans les régions boréales. Compte tenu de l'importance de la forêt boréale pour de nombreux oiseaux migrateurs, nous avons entrepris (1) de résumer les principaux défis associés à la surveillance des populations aviaires, (2) de décrire les outils statistiques disponibles pour la surveillance des populations et leurs applications et (3) d'identifier les orientations futures afin de surmonter les difficultés actuelles de surveillance des populations aviaires dans la forêt boréale. La définition et la délimitation des populations d'intérêt et l'identification des éléments qui affectent ces populations représentent les principaux défis actuellement. Cela est dû au fait que les oiseaux migrateurs peuvent être affectés par de nombreux processus limitant les populations Avian Conservation and Ecology 14(2): 8 http://www.ace-eco.org/vol14/iss2/art8/ à différents stades de leur cycle de vie annuel. Ces facteurs sont souvent structurés de manière hiérarchique et peuvent influencer les populations aux niveaux local, régional ou continental. Certains des défis associés à la délimitation des populations et l'identification des facteurs qui influencent les populations peuvent être traités à l'aide de la multitude de méthodes d'échantillonnage et d'analyse disponibles pour examiner l'évolution de la population au fil du temps. Le choix de méthodes d'analyse appropriées dépend des objectifs de l'étude et de la nature des données, par exemple des populations uniques ou multiples, les enquêtes répétées ou basées sur des comptes ou les taux démographiques. Les progrès récents des modèles de populations hiérarchiques et intégrés ont fait de certaines de ces approches analytiques les orientations les plus prometteuses pour le développement des méthodes futures. Toutefois, ces outils requièrent d'importants jeux de données ; or, l'acquisition de données suffisantes sur les populations aviaires et de variables d'explication potentielles est complexe dans la forêt boréale. Si l'on veut surmonter les défis actuels à la surveillance des oiseaux dans la forêt boréale, il convient de consacrer des efforts importants à l'intégration et à la mise à disposition des données existantes. Le renforcement des efforts d'enquête portant sur des espèces multiples jouera également un rôle important. La mise en oeuvre de programmes d'échantillonnage équilibrés avec un modèle à panel rotatif pourrait équilibrer les compromis entre réplication spatiale ou temporelle moyennant un coût raisonnable. L'amélioration de l'accès aux co-variables environnementales explicites sur le plan spatial et temporel permettrait en outre d'élaborer des modèles de population mécaniques qui amélioreront notre compréhension de la dynamique des populations d'oiseaux migrateurs. Enfin, compte tenu du fait qu'il faut parfois de nombreuses décennies pour que les programmes de surveillance à long terme produisent des tendances fiables en matière de populations et que les priorités des organisations évoluent au fil du temps, nous pensons que des efforts collaboratifs contribueront à assurer la pérennité des nouveaux programmes de surveillance.
Article
Full-text available
Northern Pintail (Anas acuta; hereafter pintail) experienced a significant population decline in North America in the 1980s but did not rebound to the previous population level the way that other prairie dabbling duck species (Anas spp.) did once habitat conditions improved. Although the population decline occurred throughout the breeding range of pintails, the decline was most pronounced and sustained in Prairie Canada, i.e., southern Alberta and Saskatchewan. Thus, we estimated and examined annual survival, recovery, and harvest rates of pintails banded in Prairie Canada from 1960–2014. Annual survival rates varied by sex but were relatively high compared to those of other dabbling duck species and increased slightly over the study period to end at 0.64 ± 0.13 (SE) and 0.74 ± 0.10 for females and males, respectively. Recovery and harvest rates varied over time but generally declined in the 1980s and increased from the early 1990s until the end of the study period. There was no clear evidence that hunting bag limit restrictions affected annual survival, recovery, or harvest rates. In addition, we could find no compelling evidence that harvest mortality was substantially additive to nonharvest mortality for pintails. However, we could not definitively ascertain the effects of the restrictions, and we suggest that a trial basis of liberalized hunting bag limits would do much to improve the understanding of harvest and population dynamics of pintails and pose little risk to the population. Based on our results, we believe that measures other than harvest restrictions will likely have to be taken to elevate the pintail population to the North American Waterfowl Management Plan objective.
Article
Full-text available
Changes to weather patterns under a warming climate are complex: while warmer temperatures are expected virtually worldwide, decreased mean precipitation is expected at mid-latitudes. Migratory birds depend on broad-scale weather patterns to inform timing of movements, but may be more susceptible to local weather patterns during sedentary periods. We constructed Bayesian integrated population models (IPMs) to assess whether continental or local weather effects best explained population dynamics in an environmentally sensitive aerial insectivorous bird, the tree swallow (Tachycineta bicolor), along a transcontinental gradient from British Columbia to Saskatchewan to New York, and tested whether population dynamics were synchronous among sites. Little consistency existed among sites in the demographic rates most affecting population growth rate or in correlations among rates. Juvenile apparent survival at all sites was stable over time and greatest in New York, whereas adult apparent survival was more variable among years and sites, and greatest in British Columbia and Saskatchewan. Fledging success was greatest in Saskatchewan. Local weather conditions explained significant variation in adult survival in Saskatchewan and fledging success in New York, corroborating the hypothesis that local more than continental weather drives the population dynamics of this species and, therefore, demographic synchrony measured at three sites was limited. Nonetheless, multi-population IPMs can be a powerful tool for identifying correlated population trajectories caused by synchronous demographic rates, and can pinpoint the scale at which environmental drivers are responsible for changes. We caution against applying uniform conservation actions for populations where synchrony does not occur or is not fully understood.
Article
Full-text available
A full understanding of population dynamics depends not only on estimation of mechanistic contributions of recruitment and survival, but also knowledge about the ecological processes that drive each of these vital rates. The process of recruitment in particular may be protracted over several years, and can depend on numerous ecological complexities until sexually mature adulthood is attained. We addressed long term declines (23 breeding seasons, 1992-2014) in the per capita production of young by both Ross's geese (Chen rossii) and lesser snow geese (Chen caerulescens caerulescens) nesting at Karrak Lake in Canada's central arctic. During this period there was a contemporaneous increase from 0.4 to 1.1 million adults nesting at this colony. We evaluated whether (i) density-dependent nutritional deficiencies of pre-breeding females or (ii) phenological mismatch between peak gosling hatch and peak forage quality, inferred from NDVI on the brood-rearing areas, may have been behind decadal declines in the per capita production of goslings. We found that, in years when pre-breeding females arrived to the nesting grounds with diminished nutrient reserves, the proportional composition of young during brood-rearing was reduced for both species. Furthermore, increased mismatch between peak gosling hatch and peak forage quality contributed additively to further declines in gosling production, in addition to declines caused by delayed nesting with associated subsequent negative effects on clutch size and nest success. The degree of mismatch increased over the course of our study because of advanced vegetation phenology without a corresponding advance in goose nesting phenology. Vegetation phenology was significantly earlier in years with warm surface air temperatures measured in spring (i.e., 25 May – 30 June). We suggest that both increased phenological mismatch and reduced nutritional condition of arriving females were behind declines in population-level recruitment, leading to the recent attenuation in population growth of snow geese. This article is protected by copyright. All rights reserved.
Article
Full-text available
The authors examined the legal, scientific, and philosophical underpinnings of the North American Bird Banding Program [BBP], with emphasis on the U.S. Bird Banding Laboratory [BBL], but also considering the Canadian Bird Banding Office [BBO]. In this report, we review the value of banding data, enumerate and expand on the principles under which any modern BBP should operate, and from them derive our recommendations. These are cast into a Mission Statement, a Role and Function Statement, and a series of specific recommendations addressing five areas: (1) permitting procedures and practices; (2) operational issues; (3) data management; (4) BBL organization and staffing; and (5) implementation. Our major tenets and recommendations are as follows: banding provides valuable data for numerous scientific, management, and educational purposes, and its benefits far outweigh necessary biological and fiscal costs, especially those incurred by the BBL and BBO; because of the value of banding data for management of avian resources, including both game and nongame birds, government supper; of the program is fully justified and appropriate; all banding data, if collected to appropriate standards, are potentially valuable; there are many ways to increase the value of banding data such as by endorsing, promoting, and applying competence and/or training standards for permit issuance; promoting bander participation in well-designed projects; and by encouraging the use of banding data for meta-analytical approaches; the BBL should apply, promote, and encourage such standards, participation, and approaches; the BBP should be driven by the needs of users, including scientists and managers; all exchange of data and most communication between banders and the BBL should become electronic in the near future; the computer system at the BBL should be modernized to one designed for a true client-server relationship and storage of data in on-line relational databases; the BBL should continue to maintain high quality control and editing standards and should strive to bring all data in the database up to current standards; however, the BBL should transfer a major portion of the responsibility for editing banding data to the bander by providing software that will permit the bander to edit his/her own data electronically before submission to the BBL; the BBL should build the capacity to store additional data tied to original band records able to be pre-edited and submitted electronically, such as recapture data, appropriate data from auxiliary marking (e.g., resightings of color-marked birds), and other data that gain value when pooled from many banders (e.g., measurements); however, the BBL should only accept such data if they are collected using standardized methods and as part of an established program designed to utilize such data; now is the time to consider options for implementing a Western Hemisphere banding program, with leadership from the BBL; the Patuxent Electronic Data Processing Section should become part of the BBL; additional scientific and technical staff must be added to the BBL; an Implementation Team should be formed to expedite our recommendations, following timetables outlined in this document.
Article
Full-text available
Demographic links among fragmented populations are commonly studied as source‐sink dynamics, whereby source populations exhibit net recruitment and net emigration, while sinks suffer net mortality but enjoy net immigration. It is commonly assumed that large, persistent aggregations of individuals must be sources, but this ignores the possibility that they are sinks instead, buoyed demographically by immigration. We tested this assumption using Bayesian integrated population modelling of Greenland white‐fronted geese (Anser albifrons flavirostris) at their largest wintering site (Wexford, Ireland), combining capture–mark–recapture, census and recruitment data collected from 1982 to 2010. Management for this subspecies occurs largely on wintering areas; thus, study of source‐sink dynamics of discrete regular wintering units provides unprecedented insights into population regulation and enables identification of likely processes influencing population dynamics at Wexford and among 70 other Greenland white‐fronted goose wintering subpopulations. Using results from integrated population modelling, we parameterized an age‐structured population projection matrix to determine the contribution of movement rates (emigration and immigration), recruitment and mortality to the dynamics of the Wexford subpopulation. Survival estimates for juvenile and adult birds at Wexford and adult birds elsewhere fluctuated over the 29‐year study period, but were not identifiably different. However, per capita recruitment rates at Wexford in later years (post‐1995) were identifiably lower than in earlier years (pre‐1995). The observed persistence of the Wexford subpopulation was only possible with high rates of immigration, which exceeded emigration in each year. Thus, despite its apparent stability, Wexford has functioned as a sink over the entire study period. These results demonstrate that even large subpopulations can potentially be sinks, and that movement dynamics (e.g. immigration) among winters can dramatically obscure key processes driving subpopulation size. Further, novel population models which integrate capture–mark–recapture, census and recruitment data are essential to correctly ascribing source‐sink status and accurately informing development of site‐safeguard networks.
Technical Report
This website provides results of temporal and spatial analyses of capture-mark-recapture and constant-effort capture-rate data on 158 landbird species collected as part of the Monitoring Avian Productivity and Survivorship (MAPS) program between 1992 and 2006. The objectives of these analyses are to provide estimates of, and explore relationships among, the vital rates and demographic parameters of each of these species in order to provide hypotheses regarding the demographic drivers of temporal and spatial variation in their population dynamics, especially as these results may help inform research, management, and conservation efforts for them.
Article
1.Fecundity estimates for demographic modeling are difficult to acquire at the regional spatial scales that correspond to climate shifts, land use impacts or habitat management programmes. Yet they are important for evaluating such effects. While waterfowl managers have historically used harvest‐based age ratios to assess fecundity at continental scales, widely available age ratios from late‐summer banding (ringing) data present an underutilized opportunity to examine a regional fecundity index with broad temporal replication. 2.We used age ratios from banding data and hierarchical mixed‐effect models to examine how fecundity of five North American dabbling duck species was affected by temporal variation in hydrological cycles, intra‐ and inter‐specific density dependence and alternate prey availability, and whether those relationships were consistent across a broad geographic area. 3.Model‐estimated fecundity was within the range of traditional harvest‐based fecundity estimates for each species. Ecological covariates explained between 16 and 53% of the temporal variation in fecundity, dependent on species. Increasing wetland inundation and an indicator of vole population irruptions were consistent predictors of increasing fecundity across all species. Species exhibited mixed positive and negative responses to interspecific and intraspecific breeding pair densities hypothesized to affect nest and brood survival respectively, highlighting the importance of integrating brood survival into fecundity metrics for precocial species. 4.Declines in fecundity over time and across space at more northern latitudes may reflect stronger policies for grassland and wetland protection in the U.S. versus Canadian portions of the prairies over the time period of our study. Maintaining the capacity of less permanent basins to rehydrate in wetter periods through easement protection benefits fecundity, particularly for late‐nesting species that acquire a greater proportion of their reproductive energy on the breeding grounds. 5.Synthesis and applications. Age‐ratios derived from post‐breeding banding operations allowed us to attribute variation in waterfowl fecundity to temporal ecological variables. Effects of habitat management for waterfowl may be masked unless analysts account for this temporal variation. Post breeding‐pulse age ratios at capture could be useful as fecundity metrics in integrated population models and for evaluating population dynamics of extensively banded nongame species, especially if adjusted for capture vulnerability using within‐season recapture data. This article is protected by copyright. All rights reserved.
Article
Identifying the demographic parameters (e.g., reproduction, survival, dispersal) that most influence population dynamics can increase conservation effectiveness and enhance ecological understanding. Life table response experiments (LTRE) aim to decompose the effects of change in parameters on past demographic outcomes (e.g., population growth rates). But the vast majority of LTREs and other retrospective population analyses have focused on decomposing asymptotic population growth rates, which do not account for the dynamic interplay between population structure and vital rates that shape realized population growth rates (λt = Nt+1 /Nt ) in time-varying environments. We provide an empirical means to overcome these shortcomings by merging recently developed "transient life-table response experiments" with integrated population models (IPMs). IPMs allow for the estimation of latent population structure and other demographic parameters that are required for transient LTRE analysis, and Bayesian versions additionally allow for complete error propagation from the estimation of demographic parameters to derivations of realized population growth rates and perturbation analyses of growth rates. By integrating available monitoring data for lesser scaup over 60 years, and conducting transient LTREs on IPM estimates, we found that the contribution of juvenile female survival to long-term variation in realized population growth rates was 1.6 and 3.7 times larger than that of adult female survival and fecundity, respectively. But a persistent long-term decline in fecundity explained 92% of the decline in abundance between 1983 and 2006. In contrast, an improvement in adult female survival drove the modest recovery in lesser scaup abundance since 2006, indicating that the most important demographic drivers of lesser scaup population dynamics are temporally dynamic. In addition to resolving uncertainty about lesser scaup population dynamics, the merger of IPMs with transient LTREs will strengthen our understanding of demography for many species as we aim to conserve biodiversity during an era of non-stationary global change. This article is protected by copyright. All rights reserved.
Book
Bayesian statistics has exploded into biology and its sub-disciplines, such as ecology, over the past decade. The free software program WinBUGS and its open-source sister OpenBugs is currently the only flexible and general-purpose program available with which the average ecologist can conduct standard and non-standard Bayesian statistics. Bayesian Population Analysis Using WinBUGS goes right to the heart of the matter by providing ecologists with a comprehensive, yet concise, guide to applying WinBUGS to the types of models that they use most often: linear (LM), generalized linear (GLM), linear mixed (LMM) and generalized linear mixed models (GLMM). Comprehensive and richly-commented examples illustrate a wide range of models that are most relevant to the research of a modern population ecologist. All WinBUGS/OpenBUGS analyses are completely integrated in software R. Includes complete documentation of all R and WinBUGS code required to conduct analyses and shows all the necessary steps from having the data in a text file out of Excel to interpreting and processing the output from WinBUGS in R.
Article
Integrated population models (IPMs) provide a unified framework for simultaneously analyzing data sets of different types to estimate vital rates, population size, and dynamics; assess contributions of demographic parameters to population changes; and assess population viability. Strengths of an IPM include the ability to estimate latent parameters and improve the precision of parameter estimates. We present a hierarchical IPM that combines two broad-scale avian monitoring data sets; count data from the North American Breeding Bird Survey (BBS) and capture-recapture data from the Monitoring Avian Productivity and Survivorship (MAPS) program. These data sets are characterized by large numbers of sample sites and observers, factors capable of inducing error in the sampling and observation processes. The IPM integrates the data sets by modeling the population abundance as a first-order autoregressive function of the previous year's population abundance and vital rates. BBS counts were modeled as a log-linear function of the annual index of population abundance, observation effects (observer identity and first-survey-year), and overdispersion. Vital rates modeled included adult apparent survival, estimated from a transient Cormack-Jolly-Seber model using MAPS data, and recruitment (surviving hatched birds from the previous season + dispersing adults) estimated as a latent parameter. An assessment of the IPM demonstrated it could recover true parameter values from 200 simulated data sets. The IPM was applied to data sets (1992-2008) of two bird species, gray catbird (Dumetella carolinensis) and wood thrush (Hylocichla mustelina) in the New England/Mid-Atlantic coastal Bird Conservation Region of the USA. The gray catbird population was relatively stable (trend 0.4% yr(-1) ), while the wood thrush population nearly halved (trend -4.5% yr(-1) ) over the 17-yr study period. IPM estimates of population growth rates, adult survival, and detection and residency probabilities were similar and as precise as estimates from the stand-alone BBS and CJS models. A benefit of using the IPM was its ability to estimate the latent recruitment parameter. Annual growth rates for both species correlated more with recruitment than survival, and the relationship for wood thrush was stronger than for gray catbird. The IPM's unified modeling framework facilitates integration of these important data sets. This article is protected by copyright. All rights reserved.