Survival on the ark: life history trends in captive parrots

Abstract

Members of the order Psittaciformes (parrots and cockatoos) are among the most long-lived and endangered avian species. Comprehensive data on lifespan and breeding are critical to setting conservation priorities, parameterizing population viability models, and managing captive and wild populations. To meet these needs, we analyzed 83, 212 life history records of captive birds from the International Species Information System and calculated lifespan and breeding parameters for 260 species of parrots (71% of extant species). Species varied widely in lifespan, with larger species generally living longer than smaller ones. The highest maximum lifespan recorded was 92 years in Cacatua moluccensis, but only 11 other species had a maximum lifespan over 50 years. Our data indicate that while some captive individuals are capable of reaching extraordinary ages, median lifespans are generally shorter than widely assumed, albeit with some increase seen in birds presently held in zoos. Species that lived longer and bred later in life tended to be more threatened according to IUCN classifications. We documented several individuals of multiple species that were able to breed for more than two decades, but the majority of clades examined had much shorter active reproduction periods. Post-breeding periods were surprisingly long and in many cases surpassed the duration of active breeding. Our results demonstrate the value of the ISIS database to estimate life history data for an at-risk taxon that is difficult to study in the wild, and provide life history data that is crucial for predictive modeling of future species endangerment and proactively managing captive populations of parrots.

INTRODUCTION

Earth is facing a biodiversity crisis of enormous proportions, with extinction rates estimated to be 1,000 – 10,000 times greater than normal background rates (Brooks et al.; 2006, Wilson, 1991). Zoos and aquariums play a critical role in conserving biodiversity (Miller et al., 2004) through research, education, conservation of habitat and genomic materials, and captive breeding (Fischer & Lindenmayer, 2000; Foose & Wiese, 2006; Mace et al., 2007; Price & Soorae, 2003; Walters et al., 2010). Captive breeding maintains viable populations and in some notable cases provides the only source of individual for reintroductions (Beck et al., 1994; Seddon, Armstrong & Maloney, 2007), as with the black-footed ferret, Mustela nigripes (Biggins et al., 1999), California condor, Gymnogyps californianus (Snyder & Snyder, 1989), Przewalski’s horse, Equus caballus przewalskii (Bouman, 2000), and Arabian oryx, Oryx leucoryx (Rahbek, 1993). Another important, albeit less widely recognized, role for captive populations is to provide behavioral, physiological and life history data that are difficult, costly or time-consuming to obtain in field studies (see Ricklefs & Cadena, 2007). These data are useful for predictive modeling and management of wild populations and for setting management priorities for captive populations (Conde et al., 2011).

One important step in managing captive populations is to assess conservation priorities at the larger taxonomic scales of family or order. Zoos and aquariums have organized Taxon Advisory Groups (TAGs) to set priorities for maintaining and managing captive populations across higher-level taxa. TAGs determine which species to propagate based primarily on captive population numbers and conservation status (AZA, 2007; Hutchins, 2003; Wilkinson, 2000). The TAGs further divide species into Regional Collection Plans (RCP) which are represented worldwide. All of these programs were initiated in the 1980s to track and manage the genetics and demographics of captive animal populations in studbooks so as to meet overall management goals for that species (Hutchins & Wiese, 1991). A critical role for TAGs is to prioritize efforts across different species because both space and funding for captive animals are limited (Baker, 2007; Hutchins, 2003; Hutchins & Wiese, 1991; Smith et al., 2002).

TAGs face the issue of surplus animals, animals that have already made a genetic contribution to the program either directly or via kin (Hutchins & Wiese, 1991; Lindburg & Lindburg, 1995), and are now consuming resources that could otherwise be invested in breeding animals that would further enhance genetic diversity. To allocate limited zoo resources optimally, TAGs should work with RCPs to predict and control numbers of surplus animals (Graham, 1996; Lacy, 1995; Lindburg & Lindburg, 1995), a task which requires comprehensive data on a species’ lifespan, breeding parameters, IUCN status, current numbers, and demographics. Demographic and reproductive data are especially important for captive breeding programs (Hutchins, 2003; Hutchins & Wiese, 1991), and authors of captive-management manuals have been advised to incorporate data on lifespan and duration of active reproduction to improve breeding and reintroduction efforts (Jackson, 2003; Seddon et al., 2007). These recommendations have been followed in a few cases, most notably for elephants (Hutchins & Thompson, 2008; Wiese & Willis, 2004; Wiese & Willis, 2006). However, the comprehensive life history data needed for optimal management of captive populations are not readily available for most taxa (Baker, 2007; Hutchins & Thompson, 2008).

Here we provide comprehensive lifespan and reproductive data for the order Psittaciformes (parrots and cockatoos, hereafter ‘parrots’). The parrots are an important group in which to investigate general patterns of captive longevity and breeding. The order contains a high proportion of endangered species, with 36% of the 365 extant species of parrots (Forshaw & Knight, 2006) listed as being at risk (IUCN, 2009) and at least 18 confirmed extinctions by the end of the 20^th century, making parrots the most threatened speciose order of birds (Forshaw & Knight, 2006). They are also the longest-lived order of birds for their size (Prinzinger, 1993) with some reported lifespans exceeding 50 years (Brouwer et al., 2000). Furthermore, they are commonly held in captivity, with upwards of 20,000 parrots housed in zoos and other animal holding facilities (ISIS, 2009) and millions more held in private hands (World Parrot Trust, 2009). Successful reintroductions with captive bred parrots are challenging (Snyder et al., 1996), but feasible (Brightsmith et al., 2005; Collazo et al., 2003; Sanz & Grajal, 1998; White Jr, Collazo & Vilella, 2005). The majority of bird supplementation in the wild has come from captive breeding programs (Fischer & Lindenmayer, 2000), but these efforts are stymied by a lack of captive breeding populations for many species of high conservation concern. Instead, the current zoo population of parrots is biased toward large species that are more attractive to humans (Frynta et al., 2010). While the conventional role of zoos in the past has been entertainment (Hatchwell et al., 2007), the World Association of Zoos and Aquariums recently has asserted that the “major goal of zoos and aquariums will be to integrate all aspects of their work with conservation activities” (WAZA, 2005). Overall, the large numbers, long lifespans, and high level of endangerment of parrots results in a high burden on space in zoos and a critical need to set breeding and husbandry goals on the basis of conservation priorities.

Efforts to set conservation priorities for parrots have been hampered by a lack of life history data. While there are a few exemplary studies of life history and reproduction in wild populations (Beissinger, Wunderle Jr & Meyers, 2008; Buckland, Rowley & Williams, 1983; Heinsohn & Legge, 2003; Holdsworth and Dettmann, 2010; Koenig, 2008; Murphy, Legge & Heinsohn, 2003; Powlesland et al., 1992; Renton & Salinas-Melgoza, 2004; Rowley, 1983; Sandercock et al., 2000; Saunders, 1982), it is difficult to age adults and field studies are generally short in duration relative to lifespans. While data from captive individuals may not precisely predict lifespans in wild animals given the different stresses faced by each, a significant positive relationship between captive and wild maximum lifespans has been demonstrated generally in birds (Wasser & Sherman, 2010) as have similar rates of actuarial senescence (Ricklefs, 2000). Previous studies in parrots have provided some data on captive lifespans: Brouwer and colleagues (2000) reported maximum recorded ages for 176 species and subspecies of parrots, while Vanstreels and colleagues (Vanstreels et al., 2010) examined lifespans of confiscated wild-caught parrots in a Brazilian zoo. Neither study reported reproductive parameters. Parrot studbooks are maintained regionally and internationally, but less than 10% of all parrot species and subspecies housed in zoos are currently represented by studbooks worldwide (Bingaman Lackey, personal observation). In sum, these sources provide valuable information for some species, but there remains a pressing need for comprehensive life history data for the order as a whole.

The International Species Information System (ISIS) database contains thousands of records of parrot births, deaths, and reproduction contributed by zoos and other animal holding collections from approximately 845 member institutions in 80 countries (ISIS, 2009). This database represents a wealth of valuable information on parrots, and many other taxa, that has been largely untapped by the zoological and scientific community.

We provide a species-level analysis of lSIS records to present comprehensive life history data for parrots. We collated data from over 87,000 individuals representing over 260 species of parrots from the ISIS database to characterize lifespan and breeding parameters for each species, examine general patterns across major clades of parrot, and test the effects of mass and sex on lifespan.

METHODS

Data coding

We compiled individual lifespan records representing all available parrot species from ISIS. We used Forshaw and Knight (2006) as the taxonomic authority for common and scientific names. Our only departure from the classification of Forshaw and Knight (2006) was to elevate the three subspecies of rosellas under Platycercus elegans to full species based on (Joseph et al., 2008): the crimson rosella (P. elegans), the yellow rosella (P. flaveolus), and the orange-red and yellow rosella (P. adelaidae); otherwise we did not distinguish between subspecies. Individual birds that hatched in an ISIS facility received a HATCH date, while those that were transferred into an ISIS facility from a non-ISIS institution received an IN date. Birds transferred out of an ISIS facility received an OUT data, whereas birds that died in an ISIS facility received a DEATH date. We eliminated individuals with records that had an IN or HATCH date before the 1800s, or which were missing these dates entirely. We excluded individuals recorded as surviving less than one day from further analysis. Sorting and formatting of the data was conducted with Access 2003 (Microsoft Inc., Redmond, WA), and statistical analyses were run using JMP 8.0 (SAS Institute, Cary, NC).

Lifespan across species

To reveal trends in basic lifespan data across species, we first calculated the median lifespan and maximum lifespan for each species. Preliminary analysis indicated that many species followed a Type III survivorship curve (Ricklefs, 2008), with high initial mortality that reached an asymptote at four years of age. Thus we calculated lifespan statistics on two different datasets: a) all individuals who lived past their first day, b) individuals who survived to age four years or older. Four years exceeds the age of first reproduction for many species included in the analysis, but preliminary analyses found this age to be the best single threshold for avoiding juvenile mortality across all 260 species analyzed. We also calculated the median living adult age for individuals that were still alive as of March 24^th, 2008 as a measure of lifespan for the currently-living captive population.

We calculated the median instead of the mean as an indicator of central tendencies because the lifespan data was non-normally distributed and exhibited a positively skewed unimodal distribution (Zar, 1999). While we report these summary statistics for all species, for the purpose of statistical tests of life history relationships we excluded species with fewer than 20 individual records to increase reliability of the data and ensure that general trends would not be distorted by a few aberrant individuals. We tested the effect of sex on the maximum and median lifespan by performing the nonparametric Wilcoxon signed-rank test, which treated the two sexes of each species as a paired comparison (Zar, 1999). We then examined the relationship between body size and lifespan with least-squares regressions of log of mass versus log of maximum lifespan, median adult lifespan, and median adult age. Least-squares regressions of maximum lifespan versus median adult lifespan and median adult age were performed on log transformed data. Positive residuals from these regressions indicated species with a single individual, represented by the maximum lifespan, that lived substantially longer than their conspecifics, represented by median adult lifespan or age; negative residuals indicated species with a median adult lifespan or age that was closer to the maximum lifespan within that species.

Lifespan trends for clades

In addition to the summary statistics described across species, we examined data for species within selected clades of particular interest to zoos and captive population managers. These clades were i) Cacatua and allies (Cacatua, Callocephalon, Eolophus), ii) Trichoglossus and allies (Chalcopsitta, Eos, Trichoglossus), iii) Platycercus and allies (Barnardius, Platycercus, Psephotus, Purpureicephalus), iv) Ara and allies (Ara, Orthopsittaca, Propyrrhura), v) Aratinga, and vi) Amazona. We again excluded species that had fewer than 20 individual records from these analyses. A generalized linear model (GLM) was performed to test for the joint effects of mass and clade on the means of maximum lifespan, median adult lifespan, and median adult age and Tukey-Kramer HSD was used for post-hoc comparison between pairs of clades.

Breeding parameters

To describe breeding parameters for each species, we analyzed ISIS breeding information for female parrots. Males were not included in this analysis because paternity could not be unambiguously determined. For these analyses we excluded species with fewer than five individuals to maintain an adequate sample size while minimizing the effect of aberrant individuals. Several other types of exclusions were performed on the breeding data to balance maximizing the number of records available for analysis with maintaining accurate and conservative estimates of reproductive parameters (summarized in Table 1). We calculated medians of the age of first breeding, age of last breeding, duration of active breeding, and duration of post-breeding. Values for age of last breeding are conservative estimates as birds with an IN date were treated as newly hatched birds. Values of the post-breeding period are also conservative because we treated individuals transferred out of the system as deaths. A one-way ANOVA was performed to test for differences in the means of these four breeding parameters among the six major clades.

Table 1.

Criteria for calculation of breeding parameters

	Parameters used	Types of individual records included in analysis
Age of first breeding	Known HATCH date, age at first breeding > 0	Birds that reproduced within an ISIS facility
Age at last breeding	Both HATCH and IN date used if age at last breeding > 0	Birds whose last reproduction was after their transfer into an ISIS facility
Duration of active reproduction	Both HATCH and IN date used if age of first reproduction > 0	Birds that reproduced within an ISIS facility, included durations = 0, where an individual reproduced only once
Duration of post-active reproduction	Both HATCH and IN date used if age at last breeding > 0, both DEATH and OUT date used	Birds whose last reproduction was after their transfer into an ISIS facility, included durations = 0, where an individual died on the day of last reproduction

Conservation status

To determine whether conservation status is associated with particular demographic parameters, one-way ANOVAs were conducted to test for an effect of IUCN status on lifespan (maximum and median adult) and median breeding variables (age of first and last breeding, duration of active breeding and post-breeding).

RESULTS

Lifespan across species

We compiled 87,777 individual parrot records representing 262 species (72% of all parrot species) from ISIS. After excluding those individuals hatched prior to the 1800s or who failed to survive their first day, and species in which no individuals lived past a year, 83,212 individuals representing 260 species remained for analysis. Parrot species in captivity ranged dramatically in their maximum and median lifespans. The highest maximum lifespan recorded was 92 years for the salmon-crested cockatoo (Cacatua moluccensis). Only 12 species (< 5% of the 260 species) had an individual live past 50 years of age. Of all the species held in ISIS institutions, 50% never had an individual live beyond 22 years of age, and only 30% of these species had a median adult lifespan ≥ 10 years even after limiting data to individuals who survived juvenile mortality (≥ 4 years). In contrast, when only living animals were considered, 58% of species had a median age ≥ 10 years (Table 2, see Supplementary Table 1 for medians with quartiles).

Table 2.

Lifespan summary across species (in years)

Species common name	Scientific name	Lifespan: Max	Lifespan: Alla		Lifespan: Adults ≥ 4		Age: Adults ≥ 4b		IUCN statusc
			N	Median	N	Median	N	Median
Palm cockatoo	Probosciger aterrimus	39.97	522	5.29	299	11.60	144	13.30	LC
Yellow-tailed black cockatoo	Calyptorhynchus funereus	47.41	324	2.84	136	9.78	63	10.95	LC
Red-tailed black cockatoo	Calyptorhynchus banksii	55.26	354	4.20	180	11.84	82	13.02	LC
Glossy black cockatoo	Calyptorhynchus lathami	37.72	41	5.32	22	8.15	13	9.07	LC
Gang gang cockatoo	Callocephalon fimbriatum	27.57	291	2.07	96	9.73	25	13.77	LC
Galah	Eolophus roseicapilla	72.82	1889	2.00	687	9.33	249	10.08	LC
Sulphur-crested cockatoo	Cacatua galerita	72.95	1668	3.71	804	10.92	309	12.38	LC
Yellow-crested cockatoo	Cacatua sulphurea	39.97	977	4.15	497	8.90	221	11.23	CR
Blue-eyed cockatoo	Cacatua ophthalmica	37.95	84	7.78	50	11.66	21	11.60	VU
White-crested cockatoo	Cacatua alba	32.24	967	3.49	460	9.76	228	11.15	VU
Salmon-crested cockatoo	Cacatua moluccensis	92.55	1675	4.56	896	9.51	314	11.07	VU
Major Mitchell’s cockatoo	Cacatua leadbeateri	74.86	949	2.95	419	9.64	135	12.00	LC
Slender-billed corella	Cacatua tenuirostris	43.45	333	1.93	126	11.70	52	12.09	LC
Little corella	Cacatua sanguinea	44.01	455	2.45	189	11.52	56	12.49	LC
Ducorps’s corella	Cacatua ducorpsii	57.81	73	3.75	36	9.58	23	9.58	LC
Goffin’s corella	Cacatua goffini	30.39	581	3.90	289	10.65	131	13.46	NT
Red-vented corella	Cacatua haematuropygia	33.59	125	5.30	67	11.76	28	12.02	CR
Cockatiel	Nymphicus hollandicus	35.92	2850	2.53	949	7.06	241	8.98	LC
Black lory	Chalcopsitta atra	17.58	140	2.78	56	7.66	15	9.64	LC
Yellow-streaked lory	Chalcopsitta scintillata	25.01	176	2.13	60	9.48	20	9.03	LC
Brown lory	Chalcopsitta duivenbodei	26.54	248	1.66	69	8.90	30	10.07	LC
Cardinal lory	Chalcopsitta cardinalis	18.23	64	4.72	37	9.70	23	10.23	LC
Dusky lory	Pseudeos fuscata	19.59	568	2.76	239	7.34	100	7.46	LC
Black-winged lory	Eos cyanogenia	22.53	142	1.67	43	7.06	5	5.98	VU
Violet-necked lory	Eos squamata	18.23	166	2.28	58	6.57	20	6.82	LC
Blue-streaked lory	Eos reticulata	27.83	353	2.16	121	7.91	33	11.41	NT
Red and blue lory	Eos histrio	16.08	56	6.14	35	9.58	18	10.48	EN
Red lory	Eos bornea	29.48	881	2.78	380	6.92	145	7.86	LC
Blue-eared lory	Eos semilarvata	12.23	14	4.06	7	10.02	6	10.82	LC
Ornate lorikeet	Trichoglossus ornatus	18.49	143	3.27	62	6.45	30	7.91	LC
Rainbow lorikeet	Trichoglossus haematodus	37.94	6805	2.39	2424	6.47	1325	6.70	LC
Scaly-breasted lorikeet	Trichoglossus chlorolepidotus	24.17	187	2.60	72	6.62	25	6.62	LC
Olive-headed lorikeet	Trichoglossus euteles	18.74	167	2.59	63	7.40	23	8.15	LC
Yellow and green lorikeet	Trichoglossus flavoviridis	12.28	33	2.27	11	5.36			LC
Mindanao lorikeet	Trichoglossus johnstoniae	17.77	81	1.85	28	7.27	12	8.79	NT
Varied lorikeet	Psitteuteles versicolor	12.47	173	1.42	48	6.64	2	8.47	LC
Iris lorikeet	Psitteuteles iris	27.04	147	1.58	39	7.71	22	10.01	NT
Goldie’s lorikeet	Psitteuteles goldiei	24.13	605	1.61	161	7.01	31	8.90	LC
Musk lorikeet	Glossopsitta concinna	19.3	467	2.22	141	5.88	22	5.69	LC
Little lorikeet	Glossopsitta pusilla	20.11	153	1.41	34	5.27	6	9.75	LC
Purple-crowned lorikeet	Glossopsitta porphyrocephala	12.67	476	1.28	81	5.47	6	5.35	LC
Blue-crowned lorikeet	Vini australis	37.36	241	2.05	74	6.62	31	9.51	LC
Kuhl’s lorikeet	Vini kuhlii	2.21	1	2.21					EN
Blue lorikeet	Vini peruviana	21.12	91	1.21	30	11.66	1	17.95	VU
Collared lorikeet	Phigys solitarius	18.23	106	3.66	50	9.22	36	9.99	LC
Purple-bellied lory	Lorius hypoinochrous	18.13	12	10.66	11	11.09	6	11.19	LC
Black-capped lory	Lorius lory	30.83	482	3.13	219	7.56	107	7.50	LC
Purple-naped lory	Lorius domicella	19.23	181	2.79	77	7.16	29	9.00	VU
Yellow-bibbed lory	Lorius chlorocercus	13.81	134	5.27	87	8.88	47	13.81	LC
Chattering lory	Lorius garrulus	27.24	586	2.43	220	7.59	65	7.89	EN
Red-fronted lorikeet	Charmosyna rubronotata	2.34	3	2.30					LC
Red-flanked lorikeeet	Charmosyna placentis	19.59	198	1.47	56	5.91	28	6.91	LC
Fairy lorikeet	Charmosyna pulchella	20.78	71	1.43	21	6.65	3	15.58	LC
Duchess lorikeet	Charmosyna margarethae	4	4	4.00	4	4.00			NT
Josephine’s lorikeet	Charmosyna josefinae	10.2	22	3.57	10	5.26			LC
Papuan lorikeet	Charmosyna papou	22.43	552	1.38	140	6.80	33	8.23	LC
Whiskered lorikeet	Oreopsittacus arfaki	1.56	2	1.56					LC
Musschenbroek’s lorikeet	Neopsittacus musschenbroekii	26.54	73	2.34	26	7.23	4	16.89	LC
Kea	Nestor notabilis	50.53	773	2.88	339	10.49	114	10.80	VU
Kaka	Nestor meridionalis	35.46	191	2.13	64	12.55	17	12.25	EN
Pesquet’s parrot	Psittrichas fulgidus	27.63	159	4.84	87	8.29	39	9.73	VU
Orange-breasted fig parrot	Cyclopsitta gulielmitertii	12.23	1	12.23	1	12.23	1	12.23	LC
Double-eyed fig parrot	Cyclopsitta diophthalma	14.35	287	1.81	83	6.54	16	6.47	LC
Desmarest’s fig parrot	Psittaculirostris desmarestii	17.46	136	1.06	25	5.29	2	13.24	LC
Edwards’s fig parrot	Psittaculirostris edwardsii	20.4	122	1.66	36	7.34	8	9.81	LC
Salvadori’s fig parrot	Psittaculirostris salvadorii	14.89	14	3.69	7	9.73	3	9.73	VU
Guaiabero	Bolbopsittacus lunulatus	1.99	3	1.86					LC
Blue-rumped parrot	Psittinus cyanurus	15.23	28	2.60	9	6.85	2	11.04	NT
Red-cheeked parrot	Geoffroyus geoffroyi	5.54	2	3.07	1	5.54			LC
Blue-crowned racquet-tailed parrot	Prioniturus discurus	1.64	4	0.06					LC
Golden-mantled racquet-tailed parrot	Prioniturus platurus	24.65	17	4.79	9	15.78	4	24.65	LC
Eclectus parrot	Eclectus roratus	40.76	2521	1.92	949	9.00	416	10.08	LC
Great-billed parrot	Tanygnathus megalorhynchos	14.56	24	1.97	8	6.86	1	14.56	LC
Blue-naped parrot	Tanygnathus lucionensis	13.19	36	2.52	10	5.90	1	13.19	NT
Blue-backed parrot	Tanygnathus sumatranus	26.3	10	6.81	6	7.18	3	7.18	LC
Alexandrine parakeet	Psittacula eupatria	29.06	562	5.06	308	6.23	181	5.23	LC
Rose-ringed parakeet	Psittacula krameri	33.63	1719	3.58	797	7.69	274	9.58	LC
Mauritius parakeet	Psittacula echo	12.35	6	6.72	4	7.63	2	10.43	EN
Malabar parakeet	Psittacula columboides	14.08	15	5.18	8	7.24			LC
Emerald-collared parakeet	Psittacula calthorpae	10.59	2	10.41	2	10.41			LC
Plum-headed parakeet	Psittacula cyanocephala	19.86	410	4.22	209	5.62	117	5.23	LC
Blossom-headed parakeet	Psittacula roseata	17.15	21	6.07	13	9.07	6	10.12	LC
Slaty-headed parakeet	Psittacula himalayana	18.46	19	6.82	13	7.77	11	7.77	LC
Derbyan parakeet	Psittacula derbiana	28.22	408	2.24	167	9.07	62	9.81	LC
Nicobar parakeet	Psittacula caniceps	6.54	14	0.46	4	5.79			NT
Red-breasted parakeet	Psittacula alexandri	20.24	282	3.03	119	8.13	47	10.23	LC
Long-tailed parakeet	Psittacula longicauda	13.05	55	1.68	12	8.83	7	10.23	NT
Gray-headed lovebird	Agapornis canus	16.01	120	2.79	45	7.12	20	7.12	LC
Red-faced lovebird	Agapornis pullarius	19.23	123	4.22	62	7.55	13	7.73	LC
Black-winged lovebird	Agapornis taranta	15.82	164	2.38	59	7.01	2	6.16	LC
Peach-faced lovebird	Agapornis roseicollis	34.1	1943	2.33	663	6.74	120	8.23	LC
Masked lovebird	Agapornis personatus	24.24	997	1.46	282	5.73	106	5.56	LC
Fischer’s lovebird	Agapornis fischeri	32.24	1402	1.75	397	5.94	203	7.80	NT
Nyasa lovebird	Agapornis lilianae	19.2	227	3.63	108	7.44	4	6.38	NT
Black-cheeked lovebird	Agapornis nigrigenis	13.75	675	1.71	224	6.83	152	6.88	VU
Vernal hanging parrot	Loriculus vernalis	13	50	6.93	39	8.10	11	9.62	LC
Ceylon hanging parrot	Loriculus beryllinus	3.56	2	2.71					LC
Philippine hanging parrot	Loriculus philippensis	13.33	21	4.14	12	5.05	1	10.23	LC
Blue-crowned hanging parrot	Loriculus galgulus	21.25	837	2.18	253	6.33	65	9.00	LC
Maroon-rumped hanging parrot	Loriculus stigmatus	13.03	24	4.37	12	7.93			LC
Green hanging parrot	Loriculus exilis	3.45	3	3.10					NT
Yellow-throated hanging parrot	Loriculus pusillus	2.23	1	2.23					NT
Moluccan king parrot	Alisterus amboinensis	29.2	112	1.94	41	7.56	17	8.16	LC
Papuan king parrot	Alisterus chloropterus	22.35	77	6.06	45	10.26	10	11.66	LC
Australian king parrot	Alisterus scapularis	31.32	699	1.53	171	7.63	59	7.75	LC
Olive-shouldered parrot	Aprosmictus jonquillaceus	14.84	7	11.12	6	11.18			NT
Red-winged parrot	Aprosmictus erythropterus	27.3	390	1.83	137	9.03	41	10.69	LC
Superb parrot	Polytelis swainsonii	24.21	421	2.86	166	7.18	47	6.81	VU
Regent parrot	Polytelis anthopeplus	27.49	365	2.51	125	7.23	36	7.21	LC
Princess parrot	Polytelis alexandrae	23.98	614	2.55	210	7.36	76	7.45	NT
Red-capped parrot	Purpureicephalus spurius	17.9	138	1.06	29	8.00	4	12.91	LC
Mallee ringneck parrot	Barnardius barnardi	31.62	181	1.05	52	6.75	8	4.84	not available
Port Lincoln parrot	Barnardius zonarius	17.96	140	3.66	68	7.81	13	8.06	LC
Green rosella	Platycercus caledonicus	7.98	37	3.37	15	6.35	4	6.61	LC
Crimson rosella	Platycercus elegans	20.14	1010	1.03	139	6.50	36	7.53	LC
Yellow rosella	Platycercus flaveolus	13.31	59	2.50	21	6.18	1	4.23	LC
Adelaide rosella	Platycercus adelaidae	17.35	58	0.26	11	8.44	5	7.23	LC
Eastern rosella	Platycercus eximius	37.44	1302	1.61	334	6.58	93	6.81	LC
Pale-headed rosella	Platycercus adscitus	25.03	267	1.76	75	6.35	23	7.18	LC
Northern rosella	Platycercus venustus	21.57	191	2.57	69	6.81	4	7.81	LC
Western rosella	Platycercus icterotis	31.64	274	1.03	64	6.58	7	5.87	LC
Bluebonnet	Northiella haematogaster	15.4	173	1.44	40	6.06	2	10.42	LC
Red-rumped parrot	Psephotus haematonotus	17.74	792	1.93	238	5.27	107	5.23	LC
Mulga parrot	Psephotus varius	11.79	116	1.93	25	5.61	3	4.81	LC
Golden-shouldered parrot	Psephotus chrysopterygius	20.96	812	1.38	221	7.91	32	9.49	EN
Swift parrot	Lathamus discolor	14.18	184	1.26	28	6.28	11	6.24	EN
Antipodes green parakeet	Cyanoramphus unicolor	12.66	147	2.77	60	6.61	7	5.37	VU
Red-fronted parakeet	Cyanoramphus novaezelandiae	36.45	510	1.73	129	7.23	51	10.05	VU
Yellow-fronted parakeet	Cyanoramphus auriceps	16.48	171	1.30	51	6.54	20	6.96	NT
Orange-fronted parakeet	Cyanoramphus malherbi	12.28	17	1.87	4	7.56	1	12.28	CR
Horned parakeet	Eunymphicus cornutus	12.9	30	2.75	11	6.93	6	11.98	EN
Masked shining parrot	Prosopeia personata	17.73	4	11.46	4	11.46			NT
Red shining parrot	Prosopeia tabuensis	23.67	35	5.42	21	9.89			LC
Blue-winged parrot	Neophema chrysostoma	15.85	165	0.60	30	5.47			LC
Elegant parrot	Neophema elegans	14.91	207	1.72	54	6.22			LC
Rock parrot	Neophema petrophila	16.92	68	2.17	25	6.36			LC
Orange-bellied parrot	Neophema chrysogaster	13.27	426	0.81	81	6.29	47	7.23	CR
Turquoise parrot	Neophema pulchella	26.54	461	2.15	142	6.65	15	7.52	LC
Scarlet-chested parrot	Neophema splendida	25.41	1284	1.05	202	5.91	32	13.05	LC
Bourke’s parrot	Neopsephotus bourkii	19.4	511	2.11	155	6.06	43	6.56	LC
Budgerigar	Melopsittacus undulatus	18.01	4840	0.97	819	5.64	324	5.20	LC
Ground parrot	Pezoporus wallicus	1.17	1	1.17					LC
Brehm’s tiger parrot	Psittacella brehmii	1.92	1	1.92					LC
Vasa parrot	Coracopsis vasa	29.06	190	5.20	109	9.76	49	11.57	LC
Black parrot	Coracopsis nigra	37.69	132	4.89	73	8.95	22	9.90	LC
Gray parrot	Psittacus erithacus	48.26	4742	2.66	1979	8.23	882	8.75	NT
Brown-necked parrot	Poicephalus robustus	36	377	3.08	172	8.15	89	8.36	LC
Jardine’s parrot	Poicephalus gulielmi	20.2	227	4.30	118	6.81	44	9.20	LC
Brown-headed parrot	Poicephalus cryptoxanthus	16.2	169	2.27	63	6.61	30	8.58	LC
Niam-niam parrot	Poicephalus crassus	10.06	2	10.06	2	10.06	2	10.06	LC
Meyer’s parrot	Poicephalus meyeri	31.02	275	4.53	150	8.05	60	9.63	LC
Rüppell’s parrot	Poicephalus rueppellii	20.54	49	5.23	35	5.23	21	5.23	LC
Red-bellied parrot	Poicephalus rufiventris	22.23	192	1.92	65	7.49	34	7.72	LC
Senegal parrot	Poicephalus senegalus	27.16	736	3.88	361	6.60	186	5.47	LC
Hyacinth macaw	Anodorhynchus hyacinthinus	54.26	568	12.52	422	18.23	141	21.77	EN
Lear’s macaw	Anodorhynchus leari	43.57	13	14.56	11	17.86			CR
Spix’s macaw	Cyanopsitta spixii	32.22	9	6.18	7	15.18	3	20.21	CR
Blue and yellow macaw	Ara ararauna	48.52	2124	6.60	1297	12.55	273	20.78	LC
Blue-throated macaw	Ara glaucogularis	32.79	23	12.55	17	18.20	12	18.66	CR
Scarlet macaw	Ara macao	48.26	1360	8.36	896	14.59	188	21.24	LC
Green-winged macaw	Ara chloroptera	63.04	981	9.51	670	14.44	180	19.19	LC
Military macaw	Ara militaris	54.43	581	8.57	385	14.16	110	20.12	VU
Great green macaw	Ara ambigua	34.75	78	10.73	56	19.88	16	24.83	EN
Red-fronted macaw	Ara rubrogenys	36.21	241	6.48	143	13.84	42	19.77	EN
Chestnut-fronted macaw	Ara severus	39.67	135	5.73	80	10.57	13	19.80	LC
Yellow-collared macaw	Primolius auricollis	25.21	153	3.03	68	9.58	9	19.60	LC
Blue-headed macaw	Primolius couloni	13.3	2	13.30	2	13.30	2	13.30	EN
Blue-winged macaw	Primolius maracana	24.95	93	6.87	61	14.48	23	18.23	NT
Red-bellied macaw	Orthopsittaca manilata	9.72	43	2.01	16	6.89			LC
Red-shouldered macaw	Diopsittaca nobilis	22.91	157	2.51	59	11.64	15	18.98	LC
Thick-billed parrot	Rhynchopsitta pachyrhyncha	35.24	359	6.21	202	15.60	40	20.74	EN
Golden conure	Guaruba guarouba	60.9	373	4.59	190	14.10	34	21.68	EN
Blue-crowned conure	Aratinga acuticaudata	22.49	99	2.89	42	8.81	4	19.23	LC
White-eyed conure	Aratinga leucophthalmus	28.45	93	2.28	28	10.15	6	23.35	LC
Green conure	Aratinga holochlora	21.06	44	0.89	8	7.96			LC
Hispaniolan conure	Aratinga chloroptera	30.84	14	18.23	14	18.23	13	18.23	VU
Cuban conure	Aratinga euops	14.71	15	1.79	6	7.63			VU
Finsch’s conure	Aratinga finschi	14.16	1	14.16	1	14.16			LC
Red-fronted conure	Aratinga wagleri	24.86	35	8.32	30	8.32	2	15.58	LC
Mitred conure	Aratinga mitrata	27.83	93	3.45	41	8.44	11	27.83	LC
Red-masked conure	Aratinga erythrogenys	26.64	108	4.40	56	8.93	2	21.02	NT
Golden-capped conure	Aratinga auricapillus	18.85	25	6.73	19	8.43	1	18.85	NT
Jandaya conure	Aratinga jandaya	22.24	165	1.92	69	9.53	7	19.98	LC
Sun conure	Aratinga solstitialis	29.7	529	2.07	200	10.98	23	18.98	EN
Dusky-headed conure	Aratinga weddellii	24.57	12	5.39	8	9.27	1	24.57	LC
Brown-throated conure	Aratinga pertinax	19.73	31	1.42	10	8.58	1	13.73	LC
Olive-throated conure	Aratinga nana	7.12	5	2.09	1	7.12			LC
Orange-fronted conure	Aratinga canicularis	28.62	131	0.87	23	8.80	2	23.42	LC
Peach-fronted conure	Aratinga aurea	15.34	124	2.19	25	6.58			LC
Cactus conure	Aratinga cactorum	9.08	15	6.26	10	6.47			LC
Nanday conure	Nandayus nenday	30.24	411	3.39	185	9.00	19	20.79	LC
Patagonian conure	Cyanoliseus patagonus	34.12	439	7.11	294	11.19	62	18.56	LC
Monk parakeet	Myiopsitta monachus	24.78	455	3.76	217	6.65	5	22.24	LC
Slender-billed conure	Enicognathus leptorhynchus	20.17	23	3.87	11	9.28			LC
Maroon-bellied conure	Pyrrhura frontalis	17.42	40	1.52	13	11.05			LC
Green-cheeked conure	Pyrrhura molinae	8.34	3	8.22	3	8.22			LC
Maroon-tailed conure	Pyrrhura melanura	12.05	3	1.24	1	12.05			LC
Crimson-bellied conure	Pyrrhura perlata	22.34	21	7.41	15	8.44			LC
Fiery-shouldered conure	Pyrrhura egregia	12.16	2	12.13	2	12.13			LC
White-eared conure	Pyrrhura leucotis	10.61	2	6.57	1	10.61			NT
Painted conure	Pyrrhura picta	17.75	77	0.57	12	11.26			LC
Black-capped conure	Pyrrhura rupicola	6.09	6	0.01	2	6.09			LC
Blue-throated conure	Pyrrhura cruentata	20.38	29	10.57	17	16.85	4	18.53	VU
Barred parakeet	Bolborhynchus lineola	6.03	3	3.73	1	6.03			LC
Andean parakeet	Bolborhynchus orbygnesius	9.65	28	2.24	9	7.53			LC
Mexican parrotlet	Forpus cyanopygius	12.02	35	6.37	22	8.10			LC
Green-rumped parrotlet	Forpus passerinus	11.58	22	2.42	7	4.72			LC
Blue-winged parrotlet	Forpus xanthopterygius	27.9	15	3.03	7	9.16	2	27.90	LC
Spectacled parrotlet	Forpus cospicillatus	7.12	2	4.18	1	7.12	1	7.12	LC
Pacific parrotlet	Forpus coelestis	30.38	145	1.27	28	6.07	2	24.26	LC
Yellow-faced parrotlet	Forpus xanthops	10.5	22	4.47	12	6.18			VU
Plain parakeet	Brotogeris tirica	4.21	24	1.74	3	4.21			LC
White-winged parakeet	Brotogeris versicolurus	17.84	89	1.88	29	6.82			LC
Gray-cheeked parakeet	Brotogeris pyrrhopterus	13.78	20	3.48	9	8.03			EN
Orange-chinned parakeet	Brotogeris jugularis	22.79	43	0.04	5	7.05			LC
Tui parakeet	Brotogeris sanctithomae	15.15	6	1.86	2	10.01			LC
Black-capped parrot	Pionites melanocephala	27.07	96	3.40	45	13.21	8	20.71	LC
White-bellied parrot	Pionites leucogaster	23.14	34	4.97	26	5.79	3	20.28	LC
Pileated parrot	Pionopsitta pileata	19.51	29	8.27	22	9.20	1	19.51	LC
Caica parrot	Pionopsitta caica	28.24	1	28.24	1	28.24	1	28.24	LC
Blue-headed parrot	Pionus menstruus	23.03	79	3.07	37	7.36	1	18.90	LC
Red-billed parrot	Pionus sordidus	5.7	3	2.83	1	5.70			LC
Scaly-headed parrot	Pionus maximiliani	23.86	31	8.47	21	10.25	2	16.63	LC
Plum-crowned parrot	Pionus tumultuosus	13.76	8	4.37	4	11.29			LC
White-capped parrot	Pionus seniloides	7.6	1	7.60	1	7.60			LC
White-crowned parrot	Pionus senilis	13.45	7	12.63	4	13.02			LC
Bronze-winged parrot	Pionus chalcopterus	18.98	36	2.15	13	15.23	5	18.23	LC
Dusky parrot	Pionus fuscus	9.1	16	1.50	7	7.26			LC
Yellow-billed amazon	Amazona collaria	11.61	7	7.91	6	8.10			VU
Cuban amazon	Amazona leucocephala	30.83	292	4.09	165	7.77	24	21.24	NT
Hispaniolan amazon	Amazona ventralis	29.06	117	3.61	53	6.30	1	14.15	VU
Black-billed amazon	Amazona agilis	8	2	8.00	2	8.00			VU
Puerto Rican amazon	Amazona vittata	27.23	63	1.50	28	9.03			CR
Tucuman amazon	Amazona tucumana	26.19	109	5.08	66	5.41	10	18.96	NT
Red-specacled amazon	Amazona pretrei	24.23	18	1.26	5	23.54	4	23.69	VU
White-fronted amazon	Amazona albifrons	35.24	133	3.76	64	8.76	9	20.05	LC
Yellow-lored amazon	Amazona xantholora	22.05	11	12.33	10	12.33	1	22.05	LC
Green-cheeked amazon	Amazona viridigenalis	39.45	311	5.24	180	11.59	33	20.68	EN
Lilac-crowned amazon	Amazona finschi	30.73	133	4.47	74	11.13	15	25.29	VU
Red-lored amazon	Amazona autumnalis	37.84	192	7.28	127	11.75	26	18.44	LC
Festive amazon	Amazona festiva	30.31	35	9.66	28	18.11	11	21.10	LC
Red-tailed amazon	Amazona brasiliensis	23.2	16	18.89	15	21.93	8	22.34	VU
Red-crowned amazon	Amazona rhodocorytha	26.94	39	14.75	32	15.37	11	24.25	EN
Blue-cheeked-amazon	Amazona dufresniana	22.42	15	8.43	15	8.43	6	19.48	NT
Yellow-shouldered amazon	Amazona barbadensis	34.62	46	8.77	38	10.45	8	9.46	VU
Yellow-faced amazon	Amazona xanthops	24.73	28	6.92	23	11.90	4	24.73	NT
Blue-fronted amazon	Amazona aestiva	37.44	646	5.37	395	10.22	73	19.32	LC
Yellow-crowned amazon	Amazona ochrocephala	34.45	401	5.18	240	10.46	50	17.75	LC
Yellow-naped amazon	Amazona auropalliata	66.81	185	7.91	124	11.45	38	19.48	LC
Yellow-headed amazon	Amazona oratrix	41.5	620	3.37	285	10.48	54	20.78	EN
Orange-winged amazon	Amazona amazonica	39.02	289	5.43	174	8.44	30	18.15	LC
Scaly-naped amazon	Amazona mercenaria	6.57	2	4.52	1	6.57			LC
Mealy amazon	Amazona farinosa	42.9	100	6.11	62	10.40	11	16.93	LC
Vinaceous amazon	Amazona vinacea	26.94	43	4.28	22	12.15	6	21.65	VU
St. Vincent amazon	Amazona guildingii	36.58	32	17.30	29	19.87	3	34.70	VU
St. Lucia amazon	Amazona versicolor	38.75	27	13.57	17	19.92	7	32.04	VU
Red-necked amazon	Amazona arausiaca	29.65	3	29.61	2	29.63			VU
Hawk-headed parrot	Deroptyus accipitrinus	38.25	163	6.50	101	13.57	15	21.30	LC
Purple-bellied parrot	Triclaria malachitacea	15.14	12	5.80	9	7.99			NT
Carolina parakeet	Conuropsis carolinensis	15.55	4	10.02	3	15.03			EX

^aDoes not include chicks that died day of hatch.

^bIndividuals that were still alive as of 3/24/2008.

^c2009 Red List status: LC = Least Concern, NT = Near Threatened, VU = Vulnerable, EN = Endangered, CE = Critically Endangered, EX = Extinct

When further excluding species that do not have at least 20 individual records, 82,777 individuals from 199 species remained in the dataset used for the following lifespan analyses. Matched pairs analysis indicated that living adult median age is significantly different than median adult lifespan (age = 11.12 ± 4.79, lifespan = 8.81 ± 2.62; z₉₇ = 2036.00, P < 0.0001) with living adults today surviving longer on average than adults collectively over the last 200 years. Least-squares regressions of log maximum lifespan and log median adult lifespan on log body mass revealed that mass was a significant predictor of maximum lifespan (F_{1, 195} = 148.79, P < 0.0001, R² = 0.43; Fig. 1) median adult lifespan (F_{1, 167} = 182.65, P < 0.0001, R² = 0.52 and median adult age (F_{1, 96} = 71.11, P < 0.0001, R² = 0.43; Supplementary Fig. 1) such that larger species were generally longer lived than smaller species.

Figure 1.

The regression of log of maximum lifespan on log of adult mass by parrot species with n ≥ 20 individuals. The slope of the fit line is 0.3215 ± 0.0213.

Matched pairs analysis revealed that sex affected maximum lifespan (males = 24.79 ± 12.58, females = 23.05 ± 12.52; z₁₉₆ = 3027.00, P < 0.0001), median adult lifespan (males = 9.12 ± 3.19, females = 8.93 ± 3.09; z₁₆₈ = 1671.50, P = 0.0049), and median living adult age (males = 11.37 ± 5.00, females = 11.15 ± 4.91.; z₉₇ = 488.00, P = 0.0490) with males living longer on average (Fig. 2).

Figure 2.

Histograms of lifespan by parrot species with n ≥ 20 individuals of female (a) maximum and (b) median adult lifespan and (c) median adult age, and of male (d) maximum and (e) median adult lifespan and (f) median adult age.

Least-squares regressions of log maximum lifespan on log median adult lifespan and log median adult age revealed that both were significant predictors of maximum lifespan (lifespan: F_{1, 167} = 78.56, P < 0.0001, R² = 0.32; age: F_{1, 96} = 42.75, P < 0.0001, R² = 0.31; Fig. 3). For both regressions the cockatoos were the most notable positive outliers.

Figure 3.

Regression of log of maximum lifespan on (a) log of median adult lifespan (slope = 0.7975 ± 0.0223) and (b) log of median adult age (slope = 0.5759 ± 0.0303). Points represent individual parrot species, plus signs indicate the species with a residual value greater than 0.3, and asterisks indicate the species with a residual value less than −0.3.

Lifespan trends for clades

Our selected clades of parrots differed in most lifespan parameters. A one-way ANOVA testing for differences among clades in the means of maximum lifespan (F_{5, 77} = 13.27, P < 0.0001), median adult lifespan (F_{5, 69} = 15.65, P < 0.0001), and median adult age (F_{5, 37} = 112.14, P < 0.0001) indicated that clades differed significantly in these parameters (Fig. 4). To assess whether these differences could be attributed to the size differences among clades, both clade and weight were run as factors in a GLM. There was no significant interaction between clade and weight for maximum lifespan (χ_{5, 71} = 5.35, P = 0.3741); when this interaction term was removed and the GLM rerun, both clade (χ_{5, 68} = 32.07, P < 0.0001), and weight (χ_{1, 68} = 21.66, P < 0.0001), had a significant effect on maximum lifespan. In contrast, there was a significant interaction for median adult lifespan (χ_{5, 63} = 16.86, P = 0.0048), but neither weight (χ_{1, 63} = 3.69, P = 0.0547) nor clade (χ_{5, 63} = 5.71, P = 0.3357) had significant effects with this interaction term in the model. There were insufficient degrees of freedom to run a GLM with an interaction for median adult age, but the effect of clade was significant (χ_{5, 36} = 101.88, P < 0.0001) while that of weight was not (χ_{1, 36} = 0.56, P = 0.4544). The Cacatua clade (cockatoos) showed the greatest mean of maximum lifespan at 50.78 years. In contrast, the Ara clade had the highest mean of median adult lifespan at 14.31 years. Overall, the Cacatua clade included some of the longest-lived individuals in the entire database, but out of the species held in captivity, 65% of them had never had an individual live past 50 years old. Mean median adult lifespan for this clade was notably low in captivity (10.36 years), significantly less than the Ara clade, and did not differ from the Aratinga or Amazona clades whose mean maximum lifespans were 15–25 years less than that of the cockatoos. The median age of living birds is higher than the median lifespan of all birds for all six clades, but this increase is much less dramatic in the cockatoos than in Ara, Aratinga, and Amazona (Fig. 4).

Figure 4.

Mean and SE of (a) maximum lifespan, (b) median adult lifespan and (c) median adult age for major clades of parrots. Clades that do not share the same letter within the bar are significantly different based on a Tukey-Kramer HSD post hoc test.

Breeding parameters

Breeding parameters in captivity varied greatly across the 193 species for which breeding data was available (Supplementary Table 2). When restricted to species with data for ≥ five individuals, the lowest median age at first breeding was 1.10 years for the orange-bellied parrot (Neophema chrysogaster). The highest median age of last breeding was 19.75 years for the St. Vincent amazon (Amazona guildingii). The blue-eyed cockatoo (Cacatua ophthalmica) had the longest median breeding duration at 5.92 years. The longest median post-breeding duration was recorded at 5.16 years for Pesquet’s parrot (Psittrichas fulgidus); Supplementary Table 2).

A one-way ANOVA testing for differences in the means of reproduction data among the six selected clades indicated that they differed in the median age of first breeding (F_5,27 = 5.39, P = 0.0015), median age of last breeding (F_{5, 49} = 17.77, P < 0.0001), median duration of active reproduction (F_{5, 49} = 4.18, P = 0.0031), and median duration of post-reproduction (F_{5, 45} = 5.66, P = 0.0004). Notably, the mean median duration of post-reproduction was longer than the mean median duration of active reproduction for Trichoglossus, Cacatua, Amazona, and Platycercus (Fig. 5).

Figure 5.

Mean and SE of (a) median age at first breeding, (b) median age at last breeding, (c) median duration of active breeding, (d) median duration of post-breeding for major clades of parrots. Clade bars that do not share the same letter are significantly different based on a Tukey-Kramer HSD post hoc test.

Life history and IUCN status

After classifying species using the 2009 IUCN Red List, we found 68% of species were of Least Concern (LC), 10% were Near Threatened (NT), 11% were Vulnerable (VU), 7% were Endangered (EN), and 3% were Critically Endangered (CR; Table 1). One-way ANOVAs of lifespan and breeding parameters by IUCN status revealed that there was a detectable difference in adult median lifespan (F_{4, 163} = 9.00, P < 0.0001), median adult age (F_{4, 93} = 6.44, P = 0.0001), median age of last breeding (F_4,125 = 3.55, P = 0.0088), and median duration of active breeding (F_4,125 = 4.65, P = 0.0016) among the IUCN status groups. The species with VU, EN or CR (the classifications of highest threat) had greater average values for maximum lifespan, median adult lifespan, median age of last breeding, and median duration of active breeding than did species classified as LC or NT.

DISCUSSION

Parrots have a reputation for being one of the longest-lived avian taxa (Prinzinger, 1993). This analysis of 260 species of captive parrots spanning the order Psittaciformes demonstrates that even closely related clades of parrots can differ dramatically in lifespan and duration of reproduction. While a few individual parrots have lived for nearly a century, the majority of parrots in captivity did not live much beyond two decades. Even when accounting for juvenile mortality, only 30% of the 260 species had median adult lifespans ≥ 10 years. Clearly, most captive parrots are not living as long as generally thought. However, we found that lifespan in captivity appears to be increasing, as the median age of living adult birds is significantly greater than the median lifespan of all birds in the database, despite the truncating effect on lifespan of considering only living birds. This increase is likely due to advances in animal husbandry and indicates that modern zoos have improved their care and maintenance of parrots. Below we discuss these general trends and their implications for the conservation of parrots.

Life history trends in parrots

As found in a smaller analysis of parrot lifespan (Munshi-South & Wilkinson, 2006), we found that larger parrots had longer lifespans than smaller parrots. Even though parrots with a larger body mass generally lived longer than smaller bodied parrots, on average the difference in mean median adult lifespan and age was only about a decade within the six clades examined. When body mass was included as a covariate in the analysis mass and clade had a significant interaction for median adult lifespan, suggesting the effect of body mass on median adult lifespan varied with different clades. In contrast, body mass and clade independently affect maximum lifespan.

Male parrots had statistically longer maximum and median lifespans than females. However, this difference was small (1.74 yrs longer max lifespan, 0.22 yrs longer median lifespan) and may not be biologically important. In general, there is no consistent pattern of sex differences in avian lifespans; some sources report that in many species of birds, males live longer than females (Holmes et al., 2003), but other sources cite females as the sex with the typically longer lifespan (Christe, Keller & Roulin, 2006).

Breeding parameter patterns for captive female parrots vary greatly across species. Some smaller species were able to breed before they were a year old, while many larger species still bred when they were past 40 years old (Supplementary Table 2). Notably, several species also had very long post-breeding periods, and clade means of the median duration of active reproduction was similar to the median duration of the post-reproductive period. This similarity suggests that either a) parrots are not being housed in situations where they can realize their breeding potential fully, b) parrots have an unusually early reproductive senescence compared to other birds (Holmes et al., 2003), or c) female parrots have an extended lifespan in captivity relative to wild parrots and can live past the constraints on egg production (as seen in domestic quail; vom Saal, Finch & Nelson, 1994). Housing is likely influencing the breeding data as not all zoo parrots have access to a sexually mature, opposite-sex conspecific in ideal breeding conditions, but this effect cannot be teased apart from the other factors until data on opportunity to breed is also recorded. While this is not always feasible, especially in monomorphic species, it would be beneficial for zoos to enter as much of these data as possible into ISIS, so the impact of biological factors could potentially be assessed.

Caveats

While our analysis provides an important demonstration of the utility of the ISIS database for providing lifespan data on long-lived species, there are important caveats concerning the reliability of the data. The ISIS database is composed of data contributed by many different institutions that do not necessarily adhere to the same standards of accuracy, reliability, diligence, and comprehensiveness in record keeping. While we tried to eliminate clearly erroneous records during our initial compilation of the data, some questionable values remain (e.g. the age at first breeding of 0.29 years from Cyanoramphus novaezelandiae, Supplementary Table 2). Data accuracy can depend on the species, as parentage is harder to ascertain in group-living species than for species housed in pairs. In many cases, individual records may also represent an incomplete account of the entire lifespan due to transfers of animals in and out of ISIS member institutions. Overall, we suggest that the greatest care be exercised in generalizing from breeding data, as captive breeding is dependent on opportunities provided by housing arrangements and thus most subject to biases introduced by captivity. Renewed commitment of all ISIS members to record keeping protocols would improve the value of this large database for species maintenance, reproduction and conservation. A more fundamental issue is that ISIS data are from captive animals. While captive animals rarely suffer levels of predation and starvation seen in wild populations, they may experience higher rates of inbreeding, unusual social group composition and captive conditions that produce physical and psychological stress (Meehan & Mench, 2006). It is difficult to assess the relative importance of these factors, but there are some indications that lifespan data from captive animals are a generally reliable predictor of lifespan in the wild (Ricklefs, 2000; Wasser & Sherman, 2010).

Conservation Implications

This taxon-wide analysis of parrot lifespan and breeding parameters has several implications for conservation. First, survival in captivity should be taken into account when deciding which species to propagate. For example, the swift parrot (Lathamus discolor) had low residuals in the maximum lifespan on medium adult lifespan regression (Supplementary Table 3), meaning that many individuals of that species live nearly as long as the oldest surviving members. In contrast, the cockatoos had the highest residuals, and patterns in lifespan data that suggest while cockatoos have the biological potential to live for a very long time, few individuals are realizing that potential in captivity. We suggest that in the short term, zoos focus resources on propagating endangered species that fare well in captivity in order to create populations for potential reintroductions. International or regional studbooks should be created for the species that fit these requirements, which includes the swift parrot (L. discolor), golden-shouldered parrot (Psephotus chrysopterygius), and sun conure (Aratinga solstitialis). Long-term goals should include research aimed at improving husbandry and welfare so that endangered species that currently do not survive well in captivity, such as the cockatoos, can become better candidates for captive propagation programs.

Second, our data are the most comprehensive to date regarding lifespan and breeding in parrots. Such data are critical for parameterizing population viability models for wild populations. It is difficult to compare our captive data to data from wild populations, as the life history traits of interest have been studied for relatively few species over a limited scope of time in comparison to parrot lifespan and reproduction. The majority of these studies estimate survival rates or fecundity (Beissinger, Wunderle Jr & Meyers, 2008; Buckland, Rowley & Williams, 1983; Heinsohn & Legge, 2003; Koenig, 2008; Murphy, Legge & Heinsohn, 2003; Powlesland et al., 1992; Renton & Salinas-Melgoza, 2004; Rowley, 1983; Sandercock et al., 2000; Saunders, 1982;). A study on wild orange-bellied parrots reported life history measures comparable to our captive data (Holdsworth, Dettmann & Baker, in press). Furthermore, two of the general trends we detected have particular importance for the viability of wild parrot populations, namely the shorter median lifespans than generally considered, and the long periods of post-reproductive lifespans. Taken together these trends suggest that wild populations may be more vulnerable to rapid declines than previously thought.

Third, our results suggest the suitability of older individuals for captive breeding should be carefully assessed. Many species in our dataset exhibited long post-breeding durations (see Supplementary materials: Fig. 2 and Table 2); it is not clear whether this phenomenon also occurs in wild parrots or is an artifact of captive conditions. Efforts to house pairs together could potentially increase the duration of active breeding and thereby maximize their value for conservation. On the other hand, if these long post-breeding durations are generally characteristic of parrot life history, then many individuals will be surplus animals for as long a period as they were contributing active breeders. If true, this trend would put additional pressure on TAGs to refine their prioritization efforts.

Fourth, these data on average lifespan and breeding parameters may be used by TAGs as a rough guide for predicting future endangerment of species and proactively planning captive management priorities. We found that larger-bodied species that lived longer and bred later in life tended to be more threatened according to IUCN classifications. These trends suggest that TAGs should add lifespan and breeding measures to their existing criteria of number of individuals in captivity and IUCN status (AZA, 2007) in prioritizing the management of captive parrot populations for conservation.

Finally, this study demonstrates the general value and utility of the ISIS database and provides a baseline for demographic comparisons with wild populations. Even though caution must be exercised, ISIS provides a tremendous source of unrivaled information which can be used to parameterize population viability models for wild populations and adaptively manage captive populations according to conservation priorities.