Food resources and reproductive output of the Austral Parakeet(Enicognathus ferrugineus) in forests of northern Patagonia

Soledad DiacuteazAC Thomas KitzbergerA and Salvador PerisB

AINIBIOMA-CONICET and Laboratorio Ecotono CRUB Universidad Nacional del Comahue Quintral 1250(8400) Bariloche Riacuteo Negro Argentina

BDepartamento de Zoologiacutea Facultad de Biologiacutea Universidad de Salamanca 37071 Salamanca SpainCCorresponding author Present address 256 Cactus Court Boulder CO 80304 USA Email jisdiazgmailcomjisdiazyahoocomar

Abstract Assessing use of resources across resource gradients and over time is necessary for determining factors thatinfluence the natural distribution of birdsWe analysed the pattern of availability of food resources the use of food resourcesand the influence of food availability on reproduction and demography for the Austral Parakeet (Enicognathus ferrugineus)in two southern temperate forests of Argentine Patagonia The abundance of Austral Parakeets in Nothofagus pumilioforests co-dominated by Araucaria araucana was generally higher and fluctuated more than abundance of populationsin monospecific N pumilio forest This appears to be related to A araucana providing a larger but more variable foodresource Austral Parakeets responded quickly to changing availability of resources by modifying their diet if alternativefood resources were available and reproductive success also increased when seed availability was high We found mixedforests to be a unique and important habitat for Austral Parakeets in the northern part of their range and conservationmanagement should recognise this special role that mixedN pumiliondashA araucana forests play in Austral Parakeet ecology

Additional keywords Araucaria araucana habitat use Nothofagus pumilio pollen seed psitacid southern temperateforest

Received 23 September 2011 accepted 20 April 2012 published online 21 August 2012

Introduction

The availability of food affects avian reproduction by influencingthe timing of breeding initiation clutch-sizes and nestling sur-vival and thus ultimately has an effect on vital demographicrates andpopulation trends (Lack1966Martin 1987Hutto 1990Newton 1998 Ferretti et al 2005 Salafsky et al 2007 Verhulstand Nilsson 2008) Foraging theory predicts that populations ofa species with greater access to food may have greater overallbreeding success than populations in poorer environments(in terms of quantity or quality of food) (Stephens and Krebs1989) For example intra- and inter-annual variation in theavailability of food can affect female nutritional condition andthe onset of breeding which in turn can have both direct andindirect effects on the survival of nestlings (Newton 1998Salafsky et al 2007 Verhulst and Nilsson 2008 Zaacuterybnickaacute2009) Furthermore spatial variation in the availability of foodmay affect variation in the breeding success of individuals anddefensive behaviour of breeding pairs as well as overall popu-lation growth and density (Newton 1998 Fontaine et al 2009)

Very few studies have considered the influence of temporaland spatial variation in the availability of foodon the reproductionof neotropical parrots (Psittaciformes) but those that do haveshown an important influence of food supply on the growth andsurvival of nestlings (Renton 2002 Masello and Quillfeldt 2003

2004 Renton and Salinas-Melgoza 2004 Sanz and Rodriacuteguez-Ferraro 2006) Understanding the influence of environmentalfactors on breeding success and population vulnerability toenvironmental change is urgently needed for the managementand conservation of Neotropical parrots The relationship be-tween inter-annual variation in environmental conditions andavailability of food are particularly important for psittacid speciesin temperate regions where marked seasonal variations limitresource availability and affect parental condition (Masello andQuillfeldt 2003) and thus reproductive parameters breedingsuccess and population size

TheAustral Parakeet (Enicognathus ferrugineus) has themostsoutherly distribution of any psittacid being restricted to thesouthern temperate forests of Andean Patagonia (36ndash54S) (For-shaw 2010) Its distribution coincides with the distribution ofNothofagus forests (Veblenet al 1996) and in thenorthernpart ofits range with the distribution of Araucaria araucana forests(Gonzalez et al 2006)BothNothofaguspumilio andAaraucanahave strong masting patterns of seed production (Gonzalez et al2006 Heinemann 2007 Sanguinetti and Kitzberger 2008) de-fined as the intermittent and synchronous production of largeseed crops by individuals in populations of long-lived plants(Kelly 1994) Yet the cycles betweenN pumilio andA araucanaare not synchronised and vary in length (Gonzalez et al 2006

CSIRO PUBLISHING

Emu 2012 112 234ndash243httpdxdoiorg101071MU12005

Journal compilation BirdLife Australia 2012 wwwpublishcsiroaujournalsemu

Heinemann 2007) thereby producing cyclical pulses of foodresources at inter-annual to decadal time-scales that are especiallyimportant to granivores (Ostfeld andKeesing 2000 Schnurr et al2002 Areta et al 2009) The Austral Parakeet is granivorous anddepends on flowers and seeds of N pumilio growing in mono-specific forests (Diacuteaz and Kitzberger 2006) with almost 100 ofits breeding season diet composed of food derived fromN pumilio There are also anecdotal reports of Austral Parakeetsfeeding on A araucana seeds (Gonzalez et al 2006 Sanguinettiand Kitzberger 2008)

In this paper we present data on the breeding biology ofthe Austral Parakeet in mixed N pumiliondashA araucana forestand forests dominated solely by N pumilio (monospecificN pumilio forests) and we analyse the relationship betweenreproductive parameters and fluctuations in the availability offood By assessing seasonal use of resources in these two forestswe sought to clarify how Austral Parakeets use resourcesthroughout their range and how they respond to variation in theavailability of food owing to the masting cycles of their foodplants The main objectives of our study were to determine arange of reproductive parameters and the relative abundanceof Austral Parakeets to identify yearly fluctuations in theseparameters and to identify dietary factors that may beresponsible for any variation observed We hypothesised thatthe reproductive performance and relative abundance of AustralParakeets would be highly influenced by tree-species of theforests they inhabit and the differential patterns of foodavailable in each of these southern temperate forest types Wepredicted different foraging strategies in populations of AustralParakeets in mixed N pumiliondashA araucana forests comparedto populations in monospecific N pumilio forests owing todifferences in the diversity quality and year-round availabilityof food Compared to monospecific forests we expected thatmixed forests would provide a more stable food source owingto the non-synchronicty of food resources on inter-annual (dueto masting) and intra-annual (differential phenology) time-scales In turn we predict higher hatching rates and reproductivesuccess of Austral Parakeets in mixed forests than in monospe-cific forests

Methods

Study area and study species

This study was conducted in two representative forests of north-western Patagonia Argentina ~180 km apart between 2007 and2010

A monospecific forest of N pumilio (200 ha) in ChallhuacoValley in the Rio Negro Province (41150S 71160W) Theelevation of this forest ranges from 1000 to 1400m above sealevel (asl) with a transition from lowland shrubbe-steppe inthe east to N pumilio forest in the west The understory of theforest consists of the herb Alstroemeria aurea and the lowshrubs Ribes magellanicum and Berberis serrato-dentata

A mixed forest of N pumilio and A araucana (168 ha) in theTromen area in Neuqueacuten Province (39350S 71250W 1050masl) This forest is fairlyflatwith noaltitudinal gradient and theunderstorey is dominated by the herb Alstroemeria aurea andshrub Ribes magellanicum

Both forests have been affected and shaped by fire eventsalthough Challhuaco has been more affected by uncontrolledlogging Climate is similar at both sites with marked seasonsconsisting of dry summers (DecemberndashMarch) with averagedaily temperatures of 17ndash19C and periods of intensive rain andsnow between late autumn and early spring (AprilndashSeptember)with average daily temperatures of 7ndash8C Annual mean precip-itation is 1300ndash1800mm in the Challhuaco Valley (Paruelo et al1998 Barros et al 1983) and 2000ndash3500mm in the Tromen area(Autoridad Interjurisdiccional de Cuencas MalalcoMeteorolog-ical Station data available by request)

Austral Parakeets nest only in tree-cavities mainly in Notho-fagus species Breeding occurs once per year betweenDecemberand March and the start of nesting is highly synchronisedbetween nesting pairs living in the same forest Broods are of5ndash8 young sometimes as many as 11 and both parents help raisethe young (Diacuteaz in press) Broods are large compared with thoseof other Neotropical Psittacidae (Masello and Quillfeldt 2002Renton and Salinas-Melgoza 2004 Brightsmith 2005)

Phenology of food items

Over the 3 years 2007ndash10 we monitored all types of food bothpotentially available and seen to be eaten by Austral Parakeets atboth sitesToassess availabilityof food (ie relative abundanceofpotentially consumable food types)we recorded the phenologicalstate of foliage and reproductive structures of all food-typesconsumed by Parakeets In each study site plant phenology wasassessed monthly along 300 m-long transects located within theforest and a minimum of 50m from roads and trails to avoid thepresence of edge species In the N pumilio forest we sampledalong 18 transects placed every 60m orientated perpendicular tothe altitudinal gradient In the mixed forest where no altitudinalgradient was evident we sampled along 20 randomly placedtransects Along each transect ten 10 10-mplotswere placed at20-m intervals starting at 10m (n= 10 plots per transect) Withineach plot we visually assessed the number of all trees shrubsherbs and hemiparasitic plants that may be food for AustralParakeets Each transect was considered one sampling unit sothe value of each transect was recorded as the average of allplots along the transect To record phenological stage monthlysurveys of presence of flowers and fruit were conducted usingbinoculars (Zeiss 8 30 Conquest T Germany) for all speciespresentwithin transect plots Fruits and seedswere included in theanalysis once they were ripe enough to be eaten indicated bycolour and size (depending on species) following previousobservations of Parakeets feeding on them (S Diacuteaz pers obs)Only reproductively mature trees of N pumilio and A araucanawere recordedMature treeswere defined as thosewith a diameterat breast height (DBH) gt20 cm and taller than 5m forA araucana (Sanguinetti and Kitzberger 2008 S Diacuteaz persobs)

Flower and seed production

During the study period and in both study sites seeds and maleflowers of N pumilio were quantified once mature Nothofaguspumilio is monoecious so male and female flowers were countedon the same plants in October and February respectively between2007 and 2010 A single branch 30 cm long was cut from 20

The diet of Austral Parakeets in Patagonia Emu 235

N pumilio trees to count the number of flowers and seeds(40 branches total) trees were randomly selected and were notfrom the phenological transects A aruacana is dioecious Pollencones of A araucaria were counted when mature in Novemberbetween 2007 and 2010 Data on production of A araucana seedcones were obtained for a broader range of years from 2005 to2010 and were counted during January A araucana reproduc-tive structures were counted (using 8 30 binoculars as above)on 72 randomly selected trees of each sex The total number ofcones per tree was estimated by counting at least half of the entiretree depending on the amount of visible crown and then mul-tiplying the number of counted cones by the percentagemeasuredto estimate the uncounted cones The study period includedone marked A araucana masting year (2007) and two non-masting years

To determine annual pollen production pollen was collectedfrom 100 male N pumilio flowers from 10 different N pumiliotrees and 20 male A araucana cones from 20 differentA araucana trees (all randomly selected) Pollen from eachflower or cone was weighed (using a 00001-g precision Metterscale) and used to calculate the mean quantity of pollen producedper flower or cone Total pollen production per tree wascalculated as the number of flowers or cones per treemeanweight of pollen per cone or flower (g) Average annual pollenproduction for each species was estimated as the average totalpollen production for all 20 N pumilio trees for which flowerswere counted and the 72 A aruacana trees for which cones werecounted

Seed production of each N pumilio was calculated as thenumber of seeds per cut branch the number ofmain branches onthe same tree As seeds were found only in the outer part of mainbranches this method may be an appropriate relative measurealthough it may be an underestimation Seed production forA araucana was estimated by multiplying the average numberof seeds per cone produced that year (Sanguinetti and Kitzberger2008) by the number of female cones counted on each of the 72trees measured In this way the mean annual production of conesand seeds was obtained for each year for each species

Seeds that fall from A araucana in autumn (MarchndashJune)often remain intact until the following spring (SeptemberndashDecember) and may serve as an important source of pre-repro-ductive food forAustral ParakeetsCurrently there is no evidenceof any bird other than the Austral Parakeet foraging on fallenA araucana seeds from the crop of the previous years althoughwild Boar (Sus scrofa) and rodents frequently eat them(Sanguinetti and Kitzberger 2008) However noticeablydifferent markings on the seed husks left by wild Boars rodentsand Austral Parakeets allow clear distinction between seed pre-dators (S Diacuteaz pers obs) During OctoberndashDecember ofeach year we counted fallen but uneaten A araucana seeds ina transect 2 20m (based on the maximum A araucanaseed dispersal distance of up to 20m from an individual treeGonzalez et al 2006) extending from the base of 20 differentfemale trees Transects began at the trunk of the tree and wereassigned a random orientation in such a manner that no overlapoccurred between neighbouring female A araucana trees Avail-ability ofAaraucana seeds from theprevious seasonon the forestfloor was included in the spring analysis of food availability(see Statistical analysis below)

Diet

Dietwas assessed by direct observation Foraging Parakeets wereobserved by walking systematically through the study sitesbetween 0800 and 1100 hours four observers were used in pairsObservations were recorded daily during the first and third weeksof every month between October and March and 4 daysper month for the rest of the year (total observation time = 720observation hours in mixed N pumiliondashA araucana forest and588 h inmonospecificN pumilio forest)When aflock (2 birds)was detected foraging one bout was recorded (thus ensuring theindependence of each observation) as well as the location typeof item consumed and the species of plant involved (Galetti1993 Walker 2007) Feeding bouts varied from a few secondsto several minutes A recorded feeding bout wasmarked as endedwhen the flock stops eating (and leaves the area)

Previous observations of the species suggest that all individualParakeets in a flock foraged on the same food item at that time(S Diacuteaz pers obs)

Reproduction of Austral Parakeets

Breeding was studied at 24 natural nests in the N pumilio forestand 39 nests in the mixedN pumiliondashA araucana forest over thethree breeding seasons (DecemberndashMarch) of the study (2007ndash08to 2009ndash10) All nests in theN pumilio forest and 89of nests inthe mixed forest were in N pumilio trees with the remainingmixed-forest nests in A araucana trees The number of eggs ornestlings were determined by inspection of nests using a home-made wireless inspection camera attached to a telescopic pole orby climbing to cavities with ropes and aluminum ladders Thecamera (6 2 2 cm)was attached to a20cm-longflexiblemetalsupport mounted on a 15 m-long pole (in two sections of 75mthat could be used independently or together) Images from thecavity were viewed from the ground by using a wireless monitorWhen climbing nest-trees cavities were examined with a mirrorand flashlight Nest-inspections were performed every other day(starting from the end of November) to determine the date oflaying until the clutch was complete (ie as Austral Parakeetslay eggs every 1ndash2 days when no new eggs had been laid after1 week) After this inspections occurred once per week until theend of the nesting period Inspections were always donewhen thefemale was outside the nest or when all adults away from the areaof thenest-treeReproductive parameters determinedwere date oflaying clutch-size hatching success (number of hatched eggs)and number of nestlings Reproductive success for each nest wascalculated as the ratio of the number of fledglings to clutch-sizeand the reproductive success of individual nests was calculatedfor both sites in each year of the study No nests failed during thestudy period (all fledged at least one young)

Abundance of Austral Parakeets

Relative abundance was determined by counting individuals inflocks (aswell as number offlocks) at social siteswhere Parakeetscongregated during the earlymorning (S Diacuteaz pers obs) two inthe N pumilio forest and three in the mixed forest Counts weremade between 0600 and 0900 hours over 4 consecutive daysin each study site in the period before the start of laying (end ofNovember roost survey method suggested by Casagrande andBeissinger 1997) At each social site two observers counted

236 Emu S Diacuteaz et al

flocks at the same time (thus four observers counted in theN pumilio forest and six in the mixed forest at the same time)For the mixed forest counts were done from 2005 to 2010whereas in the N pumilio forest counts were done only for2007ndash10 Flock counts were pooled over 1-min periods so thatthe maximum number of Parakeets observed at the same time inall social sites in each forest was considered the relative abun-dance of Parakeets for the given forest

Statistical analyses

Diet

The availability of foodwas defined as the number of availablefood types (number of species with seeds or flowers) eachmonthTemporal changes in availability were then analysed using apaired Wilcoxon test to evaluate differences in the number ofavailable food types between months A Mantel test was per-formed to determine possible associations between the availabil-ity of potential food types and their actual use throughout the yearand between study sites The Mantel test was performed usingpresence (1) ndash absence (0) matrices of available food types versusfood types that were used (1) ndash non-used (0) This method ofanalysing diet emphasises the diversity of different food typesingested by the Parakeets but does not consider the amount ofeach food that is eaten However the duration of feeding boutscould not be used as a complementary measure of diet becauseParakeets tend to fly away when they detect an observer In thisway the Austral Parakeetrsquos diet was estimated based on thefrequency of feeding bouts for each food type (Galetti 1993)

Nichebreadthwas evaluatedusing the standardisedHurlbertrsquosniche-breadth index (Krebs 1989) where values close to 0indicate dietary specialisation and values close to 10 indicatea broad diet Hurlbertrsquos index uses the Levinsrsquo index (Levins1968) and incorporates a measure of the proportional abundanceof resources Hurlbertrsquos index was calculated from observationsof Parakeet diet during OctoberndashMarch using the number ofParakeets feeding on a given food type and the number of fruitingindividuals of each food plant species recorded in plot transectseach month

Similarity or overlap in Parrot diet between the two study siteswas evaluated using theMorisita index of similarity (Krebs 1989)for quantitative data on occurrence of food types in the diet Thenumber of Parakeets observed feeding on each food type betweensiteswas used for this test This index varies from0 (no similarity)to 10 (complete similarity)

Reproduction and abundance

The number of eggs and nestlings per nest and reproductivesuccess (the ratio of the number of fledglings to the number ofeggs per nest)were tested for normality using aShapiro-Wilk testNon-parametric KruskalndashWallis tests with Dunn post hoc testsand MannndashWhitney tests were performed to determine if repro-duction differed between the N pumilio forest (n= 24) and themixed forest (n= 39) over the three years of the study The relativeabundance of Parakeets in the mixed forest was correlated withmean seed production of A araucana from the previous yearusing Pearsonrsquos correlation coefficient

All statistical analyses were performed with R statisticalsoftware version 2131 (R Development Core Team Foundation

for Statistical Computing Vienna Austria wwwr-projectorgaccessed 20March 2012)Allmeanvalues are reportedse andvariables were consider to be significant at P lt 005 unless statedotherwise

Results

Availability of food and consumption

Araucaria araucana has an average of 085 g of pollen per coneand a mean of 160 male cones per tree in a non-mast year and626 cones per tree in a mast year N pumilio has an average of001 g of pollen per flower with an average of 24 376 maleflowers per tree Individual A araucana cones produce an orderof magnitude more pollen than individual N pumilio maleflowers However based on the estimated weight of pollen percone or male flower (above) estimated production of pollen perA araucana tree was 1360ndash5321 g and 2438 g in N pumilioThemeanweight of anA araucana seed is 35 g (Sanguinetti andKitzberger 2008) and that of aN pumilio seed is 98mg (Cuevas2000) The estimated weight of seeds per A araucana tree was098 kg in non-masting years and 11875 kg in masting yearswhereas the estimated weight of N pumilio seeds per tree was02 kg

Availability of food in the forests during the breeding seasonof Austral Parakeets was not significantly different between sites(Z= 02 P = 0089 Fig 1) The asynchronous flowering andseeding phenology of N pumilio and A araucana in the mixedforest provided a more consistent supply of food throughoutthe year than in the monospecific N pumilio forest (Table 1) Incontrast the monospecific N pumilio forest showed a criticalscarcity of food at the end of autumn and in winter (AprilndashSeptember) when little more than leaf buds and fungus wereoccasionally available as food There was no such shortage in themixed forest because A araucana seeds remained in trees andon the forest floor and the phenology of N pumilio in the mixedforest wasmore advanced temporally than in theN pumilio forest(Table 1)

We observed 93 Austral Parakeet feeding bouts inthe N pumilio forest on three plant species in two genera

14 10

08

06

04

02

0

12

10

8

6

4

Num

ber

of fo

od it

ems

avai

labl

e

Hul

bert

rsquos s

tand

ardi

sed

inde

x

2

0October November December January February March

N pumilio forest

Hulbert index North Hulbert index mixed

Mixed forest

Fig 1 Number of food types available and Hulbertrsquos niche-breadthindex for monospecific N pumilio forest (Challhuaco) and mixedA araucanandashN pumilio forest (Tromen) for each month of the pre-reproductive and reproductive season of Austral Parakeets

The diet of Austral Parakeets in Patagonia Emu 237

(75 in N pumilio and 18 in Misodendrum punctulatum andM linearifolium combined these being hemiparasites ofN pumilio) and in 133 feeding bouts in the mixed forest on fourplant species in three genera (48 inN pumilio 72 in A araucanaand 13 in M punctulatum and M linearifolium combined) Themain foods consumed were male flowers and cones and seeds ofN pumilio A araucana and Misodendrum species (Table 1)Austral Parakeets used items as they became available inboth forests (Table 1) There were occasional observations ofParakeets feeding on ephemeral leaf galls of N pumilio (flocksof 5ndash80 Parakeets in 79 observations) seed galls (136 Parakeetsin 12 observations) and insects from within male cones ofA araucana (221 Parakeets in 69 observations) these itemsserved as additional but ephemeral food resources (ie availableonly for a few days and no longer than 1 week) These data werenot included in the statistical analysis of diet as the availability ofthe items was not quantified owing to their short temporalavailability

Food types selected by Parakeets were significantlycorrelated with their availability at both sites (Mantel Testr = 051 P = 004 r = 030 P= 001) However the Morisitaindices of similarity for both sites were fairly low inall months showing a clear difference between diets in the twoforests (Table 2) owing to the presence of food items fromA araucana in the mixed forest Differences between the dietsin the two forests were lowest in January and February (Morisitaindices intermediate) when Parakeets fed on seeds of N pumilioin both forests

In the breeding season Hulbertrsquos standardised niche-breadthindex (Fig 1) showed different patterns at the two study sites Theindex was fairly high for theN pumilio forest with the exceptionof November indicating that food was generally used accordingto its availability (Fig 1) In contrast in themixed forest the indexwas low during the pollen-feeding period (OctoberndashNovember)indicating selectivity and the index increased dramatically inJanuary and February when food resources were more abundant(Fig 1) A 1-month lagwas evident between the two forests withthe peak of food availability in theN pumilio forest in Decemberand in January in the mixed forest (r lagged = 085 df = 3P = 005)

Seasonal patterns of food consumption

In the N pumilio forest Parakeets restricted their diet to foodtypes from N pumilio and its hemiparasitesMisodendrum punc-tulatum and M linearifolium (the latter as occasional items seeTable 1) In another study the diet in early spring was found toconsist of poorly digestible Misodendrum pollen (Diacuteaz and

Table 1 The availability and use of food items by Austral Parakeets in the two forest sites and temporal relationship with life-history parametersof Austral Parakeets

Light grey bars indicate common presence of food item dark grey bars abundant presence X use O occasional use and X use of previous yearrsquos seed cropLife-history phenologyPL pre-laying (courtship copulation prospecting for nest-cavities preparationof nest cavity) I laying and incubationR brood-rearing

F fledging NB non-breeding (timing based S Diaz V Ojeda and A Trejo unpubl data)

Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug

Monospecific N pumilio forestFlowersNothofagus pumilio X X XMisodendrum punctulatum O OMisodendrum linearifolium O O

Seeds and fruitsNothofagus pumilio X X X XMisodendrum punctulatum O OMisodendrum linearifolium O O

Life-history phenology PL I R F NB

Mixed A araucanandashN pumilio forestFlowers and male conesAraucaria araucana X XNothofagus pumilio X X XMisodendrum punctulatum OMidodendrum linearifolium O

Seeds and fruitsAraucaria araucana X X X X X X X X XNothofagus pumilio X XMisodendrum punctulatum OMisodendrum linearifolium O

Life-history phenology PL I R F NB

Table 2 Morisita index of similarity comparing diets from themonospecific Nothofagus pumilio forest and mixed A araucanandashN pumilio forest during the reproductive season of Austral Parakeets

October November December January February March

Morisita index 0279 0265 0102 0455 0367 0023

238 Emu S Diacuteaz et al

Kitzberger 2006) By early November (late spring) after the firstflowers of N pumilio emerge at lower elevations Austral Para-keets switched entirely to highly digestible N pumilio pollen(Diacuteaz andKitzberger 2006) Parakeets continued to feedonpollenuntil late December using later flowering N pumilio at higherelevations but there is nonetheless a short period where food isscarce in late December (Table 1) During this early summershortage of food Parakeets have been reported occasionallyfeedingongalling insects (cfDiacuteaz andPeris 2011) andhoneydew(Diacuteaz and Kitzberger 2006) After the ripening of seeds of thelowest elevation N pumilio in early February the diet of AustralParakeets switched almost entirely to seeds of N pumilio withseeds ofMisodendrum eaten occasionally Parakeets continued tofeedon seedsofNpumiliountilAprilwhenmost of the seeds hadfallen Food was extremely scarce in late autumn and winter Atthis time Parakeets have been reported occasionally feeding onMisodendrum buds and Cyttaria fungal fruiting bodies (presentstudy cf Diacuteaz and Kitzberger 2006) High specialisation onN pumilio pollen in spring and N pumilio seeds in summer isreflected in lowHulbertrsquos indices duringNovember and February(Fig 1)

In contrast in the mixedN pumiliondashA araucana forest high-quality food was more evenly available throughout the yearbecause both N pumilio and A araucana were providing foodresources (Table 1) In early spring (October) thediet consistedofa combination of pollen from new N pumilio flowers andA araucana seeds remaining on the ground from the previousautumn that are exposed after snow-melt By mid- to late springwith no altitudinal gradient in the mixed forest and therefore ashorter period of N pumilio pollen availability compared to themonospecific N pumilio forest Parakeets fed principally on thematuring long-lived male A araucana cones but also foragedoccasionally on Misodendrum flowers A araucana pollenformed the main source of food from late spring to early summerso the summer gap evident in theN pumilio forest was absent By

JanuaryParakeets gradually switched fromAaraucanapollen toa diet of seeds first early maturing N pumilio andMisodendrumseeds then by March to the crop of A araucana seeds whichremained available through winter

Diet and reproductive timing

The timing of themain reproductive stages differed between sites(Table 1) Clutch completion occurred 20 days earlier in themixed forest (mean 8 December range 4ndash20 December n= 39)compared to the N pumilio forest (mean 26 December range 20Decemberndash6 January n = 24) In the N pumilio forest layingcoincided with the end of N pumilio flowering whereas in themixed forest laying coincided with the peak of A araucanapollen release (lateNovemberndashearlyDecember) In theNpumilioforest incubation coincided with the summer period of foodscarcity (lsquofood gaprsquo) whereas in the mixed forest incubationcoincidedwith the availability ofAaraucanapollen PresumablyParakeets fed their nestlings mainly N pumilio seeds in theN pumilio forest whereas in the mixed forest they first fed themwithN pumilio seeds followedwithA araucana seeds during thelast several weeks of their development in the nest These seed-based diets continued throughout the fledgling period in autumnin both forests but seeds were available all winter only in themixed forest (Table 1)

Spatial and temporal variability in demography

Clutch-size was significantly different between 2007ndash08 and2009ndash10 (Table 3) with higher values in 2007ndash08 and lowervalues in 2009ndash10 (Table 4)We found no significant differencesin clutch-size between forests in these two seasons althoughclutches tended to be larger in the N pumilio forest than in themixed forest (Table 4) In 2008ndash09 the pattern in clutch-size wasreversed with significantly larger clutches in the mixed forestthan in the N pumilio forest (Table 4)

Table 3 Variation in size of clutches and broods and reproductive success in both forestsDifferences between years were assessed with KruskalndashWallis and post hoc Dunn tests differences between sites were assessed with

MannndashWhitney tests Significant results are shown in bold (with P given in parentheses)

Clutch-size Brood-size Reproductive success

Differences between years x2 = 1047 (df = 2 P= 0005) c2 = 306 (df = 2 P= 0216) x2 = 939 (df = 2 P= 0009)08ndash09 (P= 0892) 08ndash09 (P= 0177)08ndash10 (P= 0014) 08ndash10 (P= 0006)09ndash10 (P= 0010) 09ndash10 (P= 0605)

Differences between forest types U= 5 (P= 0071) U= 3969 (P lt 0001) U= 96 (P lt 0001)

Table 4 Reproductive parameters of Austral Parakeets breeding in monospecific N pumilio forest and mixedA araucanandashN pumilio forest over the 3 years of the study

Figures are means sd Reproductive success is the ratio of the number of fledglings to the number of eggs per nest

Year N pumilio forest Mixed A araucanandashN pumilio forest2007ndash08 2008ndash09 2009ndash10 2007ndash08 2008ndash09 2009ndash10

Number of nests 7 7 10 12 20 7Clutch-size 8 plusmn 24 53 plusmn 07 61 plusmn 15 66 plusmn 08 75 plusmn 14 5 plusmn 07Brood-size 45 plusmn 13 31 plusmn 04 47 plusmn 12 43 plusmn 13 60 plusmn 15 41 plusmn 09Reproductive success 056 plusmn 007 06 plusmn 009 077 plusmn 012 065 plusmn 017 08 plusmn 012 089 plusmn 018

The diet of Austral Parakeets in Patagonia Emu 239

Kitzberger 2006) By early November (late spring) after the firstflowers of N pumilio emerge at lower elevations Austral Para-keets switched entirely to highly digestible N pumilio pollen(Diacuteaz andKitzberger 2006) Parakeets continued to feedonpollenuntil late December using later flowering N pumilio at higherelevations but there is nonetheless a short period where food isscarce in late December (Table 1) During this early summershortage of food Parakeets have been reported occasionallyfeedingongalling insects (cfDiacuteaz andPeris 2011) andhoneydew(Diacuteaz and Kitzberger 2006) After the ripening of seeds of thelowest elevation N pumilio in early February the diet of AustralParakeets switched almost entirely to seeds of N pumilio withseeds ofMisodendrum eaten occasionally Parakeets continued tofeedon seedsofNpumiliountilAprilwhenmost of the seeds hadfallen Food was extremely scarce in late autumn and winter Atthis time Parakeets have been reported occasionally feeding onMisodendrum buds and Cyttaria fungal fruiting bodies (presentstudy cf Diacuteaz and Kitzberger 2006) High specialisation onN pumilio pollen in spring and N pumilio seeds in summer isreflected in lowHulbertrsquos indices duringNovember and February(Fig 1)

In contrast in the mixedN pumiliondashA araucana forest high-quality food was more evenly available throughout the yearbecause both N pumilio and A araucana were providing foodresources (Table 1) In early spring (October) thediet consistedofa combination of pollen from new N pumilio flowers andA araucana seeds remaining on the ground from the previousautumn that are exposed after snow-melt By mid- to late springwith no altitudinal gradient in the mixed forest and therefore ashorter period of N pumilio pollen availability compared to themonospecific N pumilio forest Parakeets fed principally on thematuring long-lived male A araucana cones but also foragedoccasionally on Misodendrum flowers A araucana pollenformed the main source of food from late spring to early summerso the summer gap evident in theN pumilio forest was absent By

JanuaryParakeets gradually switched fromAaraucanapollen toa diet of seeds first early maturing N pumilio andMisodendrumseeds then by March to the crop of A araucana seeds whichremained available through winter

Diet and reproductive timing

The timing of themain reproductive stages differed between sites(Table 1) Clutch completion occurred 20 days earlier in themixed forest (mean 8 December range 4ndash20 December n= 39)compared to the N pumilio forest (mean 26 December range 20Decemberndash6 January n = 24) In the N pumilio forest layingcoincided with the end of N pumilio flowering whereas in themixed forest laying coincided with the peak of A araucanapollen release (lateNovemberndashearlyDecember) In theNpumilioforest incubation coincided with the summer period of foodscarcity (lsquofood gaprsquo) whereas in the mixed forest incubationcoincidedwith the availability ofAaraucanapollen PresumablyParakeets fed their nestlings mainly N pumilio seeds in theN pumilio forest whereas in the mixed forest they first fed themwithN pumilio seeds followedwithA araucana seeds during thelast several weeks of their development in the nest These seed-based diets continued throughout the fledgling period in autumnin both forests but seeds were available all winter only in themixed forest (Table 1)

Spatial and temporal variability in demography

Clutch-size was significantly different between 2007ndash08 and2009ndash10 (Table 3) with higher values in 2007ndash08 and lowervalues in 2009ndash10 (Table 4)We found no significant differencesin clutch-size between forests in these two seasons althoughclutches tended to be larger in the N pumilio forest than in themixed forest (Table 4) In 2008ndash09 the pattern in clutch-size wasreversed with significantly larger clutches in the mixed forestthan in the N pumilio forest (Table 4)

Table 3 Variation in size of clutches and broods and reproductive success in both forestsDifferences between years were assessed with KruskalndashWallis and post hoc Dunn tests differences between sites were assessed with

MannndashWhitney tests Significant results are shown in bold (with P given in parentheses)

Clutch-size Brood-size Reproductive success

Differences between years x2 = 1047 (df = 2 P= 0005) c2 = 306 (df = 2 P= 0216) x2 = 939 (df = 2 P= 0009)08ndash09 (P= 0892) 08ndash09 (P= 0177)08ndash10 (P= 0014) 08ndash10 (P= 0006)09ndash10 (P= 0010) 09ndash10 (P= 0605)

Differences between forest types U= 5 (P= 0071) U= 3969 (P lt 0001) U= 96 (P lt 0001)

Table 4 Reproductive parameters of Austral Parakeets breeding in monospecific N pumilio forest and mixedA araucanandashN pumilio forest over the 3 years of the study

Figures are means sd Reproductive success is the ratio of the number of fledglings to the number of eggs per nest

Year N pumilio forest Mixed A araucanandashN pumilio forest2007ndash08 2008ndash09 2009ndash10 2007ndash08 2008ndash09 2009ndash10

Number of nests 7 7 10 12 20 7Clutch-size 8 plusmn 24 53 plusmn 07 61 plusmn 15 66 plusmn 08 75 plusmn 14 5 plusmn 07Brood-size 45 plusmn 13 31 plusmn 04 47 plusmn 12 43 plusmn 13 60 plusmn 15 41 plusmn 09Reproductive success 056 plusmn 007 06 plusmn 009 077 plusmn 012 065 plusmn 017 08 plusmn 012 089 plusmn 018

The diet of Austral Parakeets in Patagonia Emu 239

In contrast to clutch-size brood-size differed between forests(Table 3) with significantly larger broods in the mixed forestthan in the N pumilio forest (Table 4) Also unlike clutch-sizewe found no differences between years although in 2009 brood-size in the mixed forest was larger than in the N pumilio forest(Table 4) The patterns of brood-size may indicate a stronginfluence of local conditions such as local availability of food(Fig 1) on hatching success in populations of Austral Parakeet(Fig 2)

Reproductive success (ratio of number of fledglings to thenumber of eggs per nest) varied between years and forests(Table 3) Mean reproductive success was significantly higherin the mixed forest for all years combined (mixed forest = 077016 N pumilio forest = 069 014 MannndashWhitney TestU= 96 P= 0043) indicating an ~10 increase in the numberof eggs reachingfledgling stage in themixed forest Reproductivesuccess was also higher in mixed forest compared to theN pumilio forest nests in every year (Table 4 Fig 2)

Themean number of Parakeets in the 200-haN pumilio forestwas 153 individuals (se 95 range 144ndash163) over the 3 years ofthe study and remained fairly constant during the 3 years (Fig 3)In contrast the estimated number of Parakeets in the 168-hamixed forest was greater and varied more between years with anoverall mean of 179 individuals (se 102 range 70ndash274 Fig 3)over the 3 years of the study (MannndashWhitney U= 1296P lt 0001)

A larger dataset of relative abundance of Parakeets andproduction ofA araucana seed cones in themixed forest between2005 and 2010 showed a significant correlation between Parakeetabundance andmean production ofA araucana seed cones in theprevious year (r= 097 df = 3 P= 0005 Fig 4)

Discussion

The overall higher relative abundance of Parakeets in the mixedforest compared with the monospecific N pumilio forest andthe larger fluctuations in numbers appear to be related to thelarger but inter-annually variable year-round resource base ofA araucana food types in the mixed forest A araucana seedsconstitute a nutritionally and quantitatively important food re-source but there are large differences in the abundance of thisresource between years (Sanguinetti and Kitzberger 2008)Despite thisA araucana seeds remain packaged in female conesfor several months until they finally fall to ground in autumn andwinter (Gonzalez et al 2006) Thus pre-dispersal seed predationby Parakeets is a low cost foraging activity because Parakeets caningest a large number of seeds without moving much within thecanopy or between crowns Parakeets also feed on any seedsthat fall to the ground (which occurs commonly as the result ofdestruction of cones by parrots) particularly when they fall onsnow-covered ground Thus as a result of their large size andpersistence on trees or on the groundA araucana seeds representa fairly constant abundant low-cost and accessible source offood to Austral Parakeets In contrast the seeds of N pumilio aresmall scattered alongbranches andwithin the crowns of trees andforest canopy remain only briefly on the trees as they maturerapidly (a few weeks Rusch 1993) fall to the ground and thenrapidly decompose and so are then unavailable as food forParakeets Parakeets have only rarely been observed feeding onground in N pumilio forests while seeds are available (S Diacuteazpers obs) Therefore years of high A araucana seed productionmay increase winter survival of Parakeets because individualnutritional status before winter is higher and because fallen seedsare available throughout the winter and early spring Non-masting years ofA araucana are a critical period of food scarcityin mixed forests that are clearly associated with reduced popula-tions and reproductive success

100

090

080

070

060

050

0402008 2009 2010

N pumilio forestMixed forest

Num

ber

of n

estli

ngs

per

egg

per

nest

Fig 2 Reproductive success of Austral Parakeets in monospecificN pumilio forest (dotted line) and the mixed A araucanandashN pumilioforest (black line) over the 3 years of the study

300

Par

akee

ts r

elat

ive

abun

danc

e

250

200

150

100

50

02008 2009 2010

N pumilio forestMixed forest

Fig 3 Mean relative abundance of Austral Parakeets at each study siteduring the three breeding seasons of the study

300 40

35

30

25

20

15

10

5

0

250

200

150

100

50

02005 2006 2007 2008 2009 2010

A a

rauc

ana

cone

r pe

r tr

ee

Mix

ed fo

rest

par

akee

ts

rela

tive

abun

danc

e

Fig 4 Mean relative abundance of Austral Parakeets (line) in the mixedA araucanandashN pumilio forest from 2005 to ndash2010 in relation to annualA araucana seed production (bars)

240 Emu S Diacuteaz et al

In a previous study in monospecific N pumilio forests DiacuteazandKitzberger (2006) showed thatAustral Parakeets had adaptedto make use of highly nutritious but hard to digest N pumiliopollen in food-limited habitats By destructively harvesting largeamounts of the short-lived N pumilio flowers along altitudinalphenological gradients Parakeets were able to extend the useof pollen through the entire spring (Diacuteaz and Kitzberger 2006)In our studywe showed that Parakeets also fed on the thick-coatedpollen of A araucana (as described in Markgraf and DrsquoAntoni1978) and by doing so gain access to a rich source of proteinduring a period of otherwise limited availability of food and thatmay be crucial to successful breeding The longevity of malecones on A araucana trees (~15ndash20 days in the tree 10ndash15 dayson the forest floor Gonzalez et al 2006) is much longer thanlongevity of flowers of N pumilio (~10 days in the tree Premoliet al 2007) In contrast to the inconspicuous male flowers ofN pumilio which are consumed exclusively on the trees Para-keets takeadvantageof thegreater longevityofAaraucanaconesby feeding on those that remain intact on the ground after peakpollination (S Diacuteaz pers obs)

Thesepatternsof resource trackingandexploitationbyAustralParakeets concur with optimal foraging theory (Stephens andKrebs 1989 Moermond 1990) which predict selection of foodresources that require less handling time or lower expenditure ofenergy per unit of resource gained Consistent with this it may bepredicted that seed predators such as parrots will track foodresources in order to exploit temporal and spatial abundances inseed production Seasonal variations in diet have been noted forIndonesian (Walker 2007) Australian (Rowley and Chapman1991) and some Neotropical (Galetti 1993 Wermundsen 1997Renton 2001) parrots Owing to the scarcity and patchiness offood resources in pure N pumilio forests Austral Parakeets areforced to move distances of hundreds of metres to several kilo-metres but remaining within pureN pumilio forests in search offood and to follow the phenological phases of N pumilio flower-ing and seeding (Diacuteaz andKitzberger 2006) In contrast foragingbehaviour in mixed N pumiliondashA araucana forests is verydifferent Because food is more evenly distributed throughoutthe year in most years with no altitudinal gradients influencingphenology and is spatially concentratedAustral Parakeets spendless time and energy in searching for and handling food in mixedforests than in N pumilio forests

Higher reproductive success in mixed forests is probably theresult of two factors (1) the availabilityofAaraucana seeds fromthe previous season during the early spring pre-laying periodparticularly in years following masting events and (2) the avail-ability of protein-rich A araucana pollen during the breedingseason to supplement the diet This highlights the importanceof the timing of the availability of food resources not just theabsolute abundance of food as a critical factor influencingreproductive success

Despite consistently higher reproductive success of AustralParakeets inmixed forests they showedmuch greater interannualvariation in relative abundance in these forests than the popula-tions from monospecific N pumilio forests which varied littlethroughout the study period We suggest that the stability ofpopulations in monospecificN pumilio forests relates to the longperiodbetweenmasting eventswhich occur approximately every10 years and between which production of flowers and seeds is

low (Cuevas 2000 Heinemann 2007) Our study period wasconducted between masting events In contrast A araucana hasmuch shorter masting cycles with masting occurring approxi-mately every 3ndash4 years (Gonzalez et al 2006) During our study2007 was such a masting year and 2008 a non-masting year Wefound the change in abundance of Parakeets in response to levelsofA araucana seed production is delayed by 1 year with relativeabundance increasing in 2008 following masting in 2007 anddeclining in 2009 following a non-masting season

Smaller populations in years following non-masting seasonsmaybeaproduct of higherwintermortality rates or reduced returnrates to breeding habitats or both although our data cannotdistinguish between these mechanisms Interestingly for thosebreeding pairs that do return after non-masting years we foundreproductive success was lower than in other years within mixedforests but remained higher than in the monospecific N pumilioforests This may indicate that availability of food types inspring such as A araucana seeds from the previous season andA araucana pollen are sufficient for successful initiation ofbreeding The fact that the relative abundance and reproductivesuccess of Austral Parakeets showed a contrasting pattern innon-masting years in the mixed forest suggests that relativeabundance may be a poor measure of habitat quality for thisspecies This is supported by several studies elsewhere that showthat a density and productivity relationship is not always the rulein avian populations (see for example review from Bock andJones 2004 Johnson 2007) and variables such as survival ratesreproductive success and rates of disturbance may be also im-portant and should be considered when estimating habitat quality(see for example Van Horne 1983 Vickery et al 1992 Peacuterot andVillard 2009)

TheAustral Parakeet is themost southerly distributed parrot inthe world and is highly dependent on forests for shelter nestinghabitat and food However different forest types differ in theavailability of these important resources not only in quantity butalso in the timing of resource availability We found AustralParakeets clearly responsed to differing resource availability incontrasting habitats Austral Parakeets showed a clear functionalresponse (Kitzberger et al 2007) to changes in the availability offood by optimising their diet to include the most abundantconcentrated and nutritional food types Further Austral Para-keets showed a numerical response (Ims 1990) in abundance inresponse to the A araucana masting cycle We found mixedforests in which N pumilio provides abundant nesting sites(Ojeda 2006 Diacuteaz in press) and A araucana provides a richand stable food source are clearly a unique and important habitatfor Austral Parakeets in the northern part of their distributionHowever forests of A araucana have a very restricted distribu-tion and are vulnerable to degradation from invasive species andhuman actions (Gonzalez et al 2006) Conservationmanagementshould recognise this special role that mixed N pumiliondashA araucana forests play in Austral Parakeet ecology and takethe necessary steps to protect them

Acknowledgements

Theauthors thankCameronNaficyandPaulEdwards for their support and theDelegacioacutendeParquesNacionales especially the staff ofLaninNationalParkFinancial support come from the Spanish MEC (project CGL2004ndash01716-

The diet of Austral Parakeets in Patagonia Emu 241

Feder) Birderrsquos Exchange IdeaWild Rufford Small Grant and a CONICET(ConsejoNacional de InvestigacionesCientiacuteficasyTeacutecnicas) doctoral grant toS Diacuteaz

References

Areta J I Bodrati A and Cockle K (2009) Specialization on GuaduaBamboo seeds by three bird species in the Atlantic Forest of ArgentinaBiotropica 41 66ndash73 doi101111j1744-7429200800458x

Barros V R Cordon V H Moyano C L Mendez R J Forquera J Cand Picio O (1983) Cartas de precipitacioacuten de la zona oeste de lasprovincias de Riacuteo Negro y Neuqueacuten Report Consejo Nacional deInvestigaciones Cientiacuteficas y Teacutecnicas Buenos Aires

Bock C A and Jones Z F (2004) Avian habitat evaluation shouldcounting birds count Frontiers in Ecology and the Environment 2403ndash410 doi1018901540-9295(2004)002[0403AHESCB]20CO2

Brightsmith D J (2005) Competition predation and nest niche shiftsamong tropical cavity nesters phylogeny and natural history evolutionof parrots (Psittaciformes) and trogons (Trogoniformes) Journal of AvianBiology 36 64ndash73 doi101111j0908-8857200503310x

Casagrande D G and Beissinger S R (1997) Evaluation of four methodsfor estimating parrot population size Condor 99 445ndash457 doi1023071369951

Cuevas J G (2000) Tree recruitment at the Nothofagus pumilio alpinetimberline in Tierra del Fuego Chile Journal of Ecology 88 840ndash855doi101046j1365-2745200000497x

Diacuteaz S Biologiacutea y conservacioacuten de la cachantildea (Enicognathus ferrugineus)El Hornero in press

Diacuteaz S and Kitzberger T (2006) High Nothofagus flower consumptionand pollen emptying in the southern South American Austral Parakeet(Enicognathus ferrugineus) Austral Ecology 31 759ndash766 doi101111j1442-9993200601637x

Diacuteaz S and Peris S (2011) Consumption of larvae by the Austral Parakeet(Enicognathus ferrugineus) Wilson Journal of Ornithology 123168ndash171 doi10167609-1901

Ferretti V Llambiacuteas P E andMartin T E (2005) Life history variation ofa neotropical thrush challenges food limitation theoryProceedings of theRoyal Society of London Series B Biological Sciences 272 769ndash773doi101098rspb20043039

Fontaine J J Decker K L Skagen S K and van Riper C III (2009)Spatial and temporal variation in climate change a birdrsquos eye viewClimatic Change 97 305ndash311 doi101007s10584-009-9644-9

Forshaw J M (2010) lsquoParrots of the Worldrsquo (Helm London)Galetti M (1993) Diet of the Scaly-headed Parrot (Pionus maximiliani) in a

semideciduous forest in southeastern Brazil Biotropica 25 419ndash425doi1023072388865

GonzalezME CorteacutesM Izquierdo FGallo L Echeverria C BekkesyS and Montaldo P (2006) Autoecologiacutea de las especies ConiacuteferasChilenas (Gimnospermas) Araucaria araucana In lsquoLas especies arboacuter-eas de los bosques templados de Chile y Argentina Autoecologiacutearsquo(Ed C Donoso Zegers) pp 36ndash53 (Marisa Cuneo Ediciones ValdiviaChile)

Heinemann K (2007) Aspectos espaciales y temporales del establecimientode Nothofagus pumilio en claros de bosques maduros en los extremos deun gradiente ambiental en el Noroeste de la Patagonia PhD ThesisUniversidadNacional del Comahue San Carlos de Bariloche RioNegroArgentina

Hutto R L (1990) Measuring the availability of food resources Studies inAvian Biology 13 20ndash28

Ims R A (1990) On the adaptative value of reproductive synchrony as apredator-swamping strategy American Naturalist 136 485ndash498doi101086285109

Johnson M D (2007) Measuring habitat quality a review Condor 109489ndash504 doi10165083471

Kelly D (1994) The evolutionary ecology of mast seeding Trends inEcology amp Evolution 9 465ndash470 doi1010160169-5347(94)90310-7

Kitzberger T Chaneton E J and Caccia F D (2007) Indirect effects ofprey swamping differential seed predation during a bamboo mastingevent Ecology 88 2541ndash2554 doi10189006-15871

KrebsC J (1989) lsquoEcologicalMethodologyrsquo (Harper andRowNewYork)Lack D (1966) lsquoPopulation Studies of Birdsrsquo (Oxford University Press

New York)Levins R (1968) lsquoEvolution in Changing Environmentsrsquo (Princeton Uni-

versity Press Princeton NJ)Markgraf V and DrsquoAntoni H L (1978) lsquoPollen Flora of Argentinarsquo

(University of Arizona Press Tucson AZ)Martin T E (1987) Food as a limit on breeding birds a life history

perspective Annual Review of Ecology and Systematics 18 453ndash487doi101146annureves18110187002321

Masello J F andQuillfeldt P (2002) Chick growth and breeding success ofthe Burrowing Parrot Condor 104 574ndash586 doi1016500010-5422(2002)104[0574CGABSO]20CO2

Masello J F and Quillfeldt P (2003) Body size body condition andornamental feathers of Burrowing Parrots variation between years andsexes assortative mating and influences on breeding success Emu 103149ndash161 doi101071MU02036

Masello J F and Quillfeldt P (2004) Consequences of La Nintildea phase ofENSO for the survival and growth of nestling Burrowing Parrots in theAtlantic coast of South America Emu 104 337ndash346 doi101071MU03039

Moermond T C (1990) A functional approach to foraging morphologybehavior and the capacity to exploit Studies in Avian Biology 13427ndash430

Newton I (1998) lsquoPopulation Limitation in Birdsrsquo (Academic Press SanDiego CA)

Ojeda V S (2006) Seleccioacuten de sitios de nidificacioacuten y biologiacuteareproductiva del carpintero gigante Campephilus magellanicus (AvesPicidae) en el noroeste de la Patagonia Argentina PhD Thesis Uni-versidad Nacional del Comahue San Carlos de Bariloche Rio NegroArgentina

Ostfeld R S and Keesing F (2000) Pulsed resources and communitydynamics of consumers in terrestrial ecosystems Trends in Ecology ampEvolution 15 232ndash237 doi101016S0169-5347(00)01862-0

Paruelo J M Beltraacuten A Jobbaacutegy E Sala O E and Golluscio R A(1998) The climate of Patagonia general patterns and controls on bioticprocesses Ecologiacutea Austral 8 85ndash101

Peacuterot A and Villard M-A (2009) Putting density back into the habitat-quality equation case study of an open-nesting forest bird ConservationBiology 23 1550ndash1557 doi101111j1523-1739200901272x

Premoli A C Raffaele E and Mathiasen P (2007) Morphological andphenological differences in Nothofagus pumilio from contrasting eleva-tions evidence from a common garden Austral Ecology 32 515ndash523doi101111j1442-9993200701720x

Renton K (2001) Lilac-crowned Parrot diet and food resource availabilityresource tracking by a parrot seed predator Condor 103 62ndash69doi1016500010-5422(2001)103[0062LCPDAF]20CO2

Renton K (2002) Influence of environmental variability on the growth ofLilac-crowned Parrot nestlings Ibis 144 331ndash339 doi101046j1474-919X200200015x

Renton K and Salinas-Melgoza A (2004) Climatic variability nestpredation and reproductive output of Lilac-crowned Parrots (Amazonafinschi) in tropical dry forest of western Mexico Auk 121 1214ndash1225doi1016420004-8038(2004)121[1214CVNPAR]20CO2

Rowley I and Chapman G (1991) The breeding biology food socialorganization demography and conservation of theMajorMitchell or PinkCockatoo Cacatua leadbeateri on the margin of the Western Australianwheatbelt Australian Journal of Zoology 39 211ndash261 doi101071ZO9910211

242 Emu S Diacuteaz et al

Rusch V E (1993) Altitudinal variation in the phenology of Nothofaguspumilio in Argentina Revista Chilena de Historia Natural 66 131ndash141

Salafsky S R Reynolds R T Noon B R and Wiens J A (2007)Reproductive response of Northern Goshawks to variable prey popula-tions Journal of Wildlife Management 71 2274ndash2283 doi1021932006-357

Sanguinetti J andKitzbergerT (2008)Patterns andmechanismsofmastingin the large-seeded southern hemisphere conifer Araucaria araucanaAustral Ecology 33 78ndash87 doi101111j1442-9993200701792x

Sanz V and Rodriacuteguez-Ferraro A (2006) Reproductive parameters andproductivity of the Yellow-shouldered Parrot on Margarita Island Vene-zuela a long-term study Condor 108 178ndash192 doi1016500010-5422(2006)108[0178RPAPOT]20CO2

Schnurr J L Ostfeld R S and Canham C D (2002) Direct and indirecteffects of masting on rodent populations and tree seed survivalOikos 96402ndash410 doi101034j1600-07062002960302x

Stephens D W and Krebs J R (1989) lsquoForaging Theory Monographs inBehavioral Ecologyrsquo (Princeton University Press Princeton NJ)

Van Horne B (1983) Density as a misleading indicator of habitat qualityJournal of Wildlife Management 47 893ndash901 doi1023073808148

Veblen T Donosos C Kitzberger T and Rebertus A J (1996) Ecologyof southernChilean andArgentineanNothofagus forests In lsquoTheEcologyand Biogeography ofNothofagus Forestsrsquo (Eds T Veblen R S Hill andJ Read) pp 293ndash319 (Yale University Press New Haven CT)

Verhulst S and Nilsson J (2008) The timing of birdsrsquo breeding seasons areviewof experiments thatmanipulated timingof breedingPhilosophicalTransactionsof theRoyal SocietyofLondonSeriesBBiological Sciences363 399ndash410 doi101098rstb20072146

Vickery PDHunterML Jr andWells JV (1992) Is density an indicatorof breeding success Auk 109 706ndash710

Walker J A (2007) Dietary specialization and fruit availability amongfrugivorous birds on Sulawesi Ibis 149 345ndash356 doi101111j1474-919X200600637x

Wermundsen T (1997) Seasonal change in the diet of the Pacific ParakeetAratinga strenua in Nicaragua Ibis 139 566ndash568 doi101111j1474-919X1997tb04674x

Zaacuterybnickaacute M (2009) Parental investment of female Tengmalmrsquos OwlsAegolius funereus correlation with varying food abundance and repro-ductive success Acta Ornithologica 44 81ndash88 doi103161000164509X464911

The diet of Austral Parakeets in Patagonia Emu 243

wwwpublishcsiroaujournalsemu