Prepared by:
Susan Koenig
Jamaica Parrot Project
20 November 1998

Jamaica, the third largest island in the Caribbean, is recognized globally as an important region of biodiversity. Due to the high levels of endemism, it is one of the two hundred most important sites in the world for plants (Heywood 1995), reptiles and amphibians (Groombridge 1982) and birds (Stattersfield et al. 1998). Jamaica, in fact, has the highest number of extant endemic bird species (28) of any Caribbean island (Raffaele et al. 1998), even though neighboring Cuba and Hispaniola are 10 and 7 times larger respectively. Invertebrate diversity is equally rich, with the island serving as home to the largest and smallest butterflies in the New World (Brown and Heineman 1972). Jamaica, unfortunately, also has the distinction of having one of the highest rates of deforestation in the world (est. 7% per annum; World Resources Institute 1998-99). Where historically nearly 97% of the island was covered with closed-forest ecosystems, some 65% of these original forests have been lost (World Resources Institute 1998-99). Of the remaining forest, only 5% has survived relatively undisturbed by humans and occurs in only the steepest or most remote, inaccessible parts of the island. At least three endemic bird species have gone extinct in the last 150 years (the Jamaican poorwill (Siphonorhis americanus) and the Jamaican Petrel (Pterodroma caribbaea) following the introduction of the mongoose (Herpestes javanicus) in 1872 and a macaw species (Ara sp.) not seen since the mid-1800s and whose cause of extinction, while unknown, was quite likely due to hunting (Lack 1976, Snyder et al. 1987)). The conservation of Jamaica's remaining native forests has reached a level of critical import as the expanding human population continues to exploit remaining resources by non-sustainable practices (Stattersfield et al. 1998). Effective management of remaining forests, such that species' persistence is ensured, will require not only identifying specific habitat features and resources that directly influence survival and reproduction (single-species approach) but also, and more important, an understanding of the consequences of those features for coexisting species and the interactions among organisms (biodiversity approach) (Martin 1989). Unfortunately, such basic, yet critical, information is lacking for nearly all of Jamaica's endemic species, a trend observed for nearly all regions of important biodiversity in the neotropics (Winker 1998).

Two species contributing to Jamaica's unique avian biodiversity are the Black-billed Parrot (Amazona agilis) and the Yellow-billed Parrot (A. collaria). Coupled with an endemic subspecies of the Olive-throated Parakeet (Aratinga nana nana), Jamaica presently hosts the richest native psittacine avifauna in the West Indies. This is notable in light of the fact that the West Indian archipelago historically hosted the highest rate of endemism for psittacids and currently claims the highest rate of extinctions in the world (Forshaw 1989, Wiley 1991). Once widespread and common, both Amazona parrot species have declined in numbers and distributions (Gosse 1847, Scott 1891-3, Danforth 1928, Lack 1976, Cruz and Gruber 1981, Varty 1991). Major populations currently are restricted to extant mid-level wet limestone forests of the John Crow and Blue Mountains and Cockpit Country. This latter region represents the stronghold of the Black-billed Parrot and the only region where both species occur sympatrically in significant numbers. Populations have been estimated between 1000-2000 each, with the Black-billed Parrot thought to be the rarer of the two (Cruz and Gruber 1981, Varty 1991). More recent and systematic surveys which address the issue of difficulties in distinguishing between species (e.g., Cook et al. 1983, Varty 1991) estimate populations of at least 4000 per species (H.A. Davis, unpubl. data). Both species are CITES II listed and possession is regulated by Jamaica's Wildlife Protection Act (1945). Cruz and Gruber (1981) identified poaching as a significant short term threat, particularly for the Yellow-billed Parrot, which is preferred of the two. Poaching continues to date, although clearly not at the alarming, near-100% rates experienced by many mainland psittacines (Collar and Juniper 1992, Martuscelli 1995). Rather, habitat loss (i.e., a quantitative reduction in native food resources, shelter from inclement weather (e.g., hurricanes) and, perhaps most important, a reduction in nesting substrates) and habitat alteration (i.e. qualitative changes in remaining resources via alteration of abiotic conditions and biotic interactions along a disturbance gradient) are identified as the most important long-term threats to the persistence of Jamaica's parrots (Wiley 1991).

In 1995 the Jamaica Parrot Project was established to: (1) more accurately delineate the ranges of both species of Amazona parrots and census populations and (2) collect baseline natural history information on breeding biology and identify factors limiting reproductive success. Due to the fact that non-excavating cavity nesters, such as Jamaica's parrots, often are limited by a quantitative scarcity of suitable nesting substrates in intensively managed forests and by the quality of nest sites in large tracts of unmanaged old-growth forests (Newton 1994), we directed our efforts to identifying those features of nest trees and surrounding habitat that would render a cavity unsuitable, minimal or optimal for reproduction for these long-lived parrots which will experience variable conditions (e.g., timing and amounts of rainfall, fluctuating predator populations) for as many as 10-15 years, their expected reproductive lifetime (see Snyder et al. 1987).

We established a field station in Windsor, Trelawny, a small community on the northern edge of Cockpit Country. Due to concerns over the perceived rarity of Black-billed Parrots, we concentrated our efforts initially to locate nesting pairs of this species. However, it became readily apparent that Black-billed Parrots were locally common and, contrary to expectations, Yellow-billed Parrots nested infrequently in the edge habitat. Whether this is due to a scarcity of large trees, a requirement of Yellow-billed Parrots reported by local persons, or other factors is presently unknown. Low sample sizes have precluded testing whether Yellow-billed Parrots require larger trees in comparison to the Black-billed Parrot, which is the slightly smaller of the two species. Because of the low sample sizes and poor nesting success, only brief information will be presented for Yellow-billed Parrots in contrast to the more complete data set for Black-billed Parrots.

Both of Jamaica's Amazon parrot species demonstrate a number of traits inherent to the order Psittaciformes and to neotropical Amazon parrots. Mated individuals maintain their pair-bonds year-round and appear to be monogamous. Pairs begin to explore and defend potential nesting cavities in early-March, a period which corresponds to the end of the winter dry season. Pairs engage in vigorous territorial counter-calling bouts both intra- and interspecifically, but never did we observe direct physical battles between birds as has been reported for species that are limited by the availability of nesting cavities (Snyder et al. 1987, Munn 1992). Only once, when we observed a Black-billed pair perched at the entrance to a cavity that Yellow-billeds were defending, did an interaction escalate to a displacement flight by the defending resident Yellow-billed Parrots. However, no physical contact ensued as the Black-billeds promptly flew from the area.

Female Black-billed Parrots initiate egg-laying in late March and early April, with a high degree of synchronicity among pairs. Eggs are laid every 48 hours and incubation begins with the laying of the first egg. Following the pattern of Psittaciformes (Bucher 1983), egg size is small (average=13 g) for adult body size and incubation (26 days) is longer than expected for egg-mass. Females alone incubate, with males providing food to their mates 3-4 times per day. Nestlings hatch asynchronously. Although incubation is prolonged, neonates are extremely altricial, with eyes and ears fused, only a few wispy down feathers, and an inability to thermoregulate. Females continue to brood nestlings continuously until the oldest chick is approximately 10-14 days old, the predicted age at which an individual nestling is able to maintain body temperature (Dunn 1975). Although the youngest chick is too young to thermoregulate, huddling siblings substitute for the brooding female. Nestling care is biparental with both parents typically entering the cavity to feed chicks. Rare was the pair where the female received food transfers from her mate at the cavity entrance and she alone entered to feed chicks. Also typical of psittacines, nestling growth rates were slow and chicks remained in the nest for 55-60 days.

Less than 1 in 3 pairs that attempt to breed produce fledglings. This is low compared to other island-dwelling Amazon parrots (e.g., 75% successful), but comparable to several mainland parrot species . Predation was the most important cause of total clutch failure. Of 61 breeding attempts, 34 (55%) resulted in the deaths of all nestlings. In cases where we had access to nest contents and could attribute causes (n=30), 21 failed due to predation. Other causes of total nest failure include: probable hypothermia (6), abandonment (2), and researcher interference while cutting access door (1). Although we do not have direct evidence of the predator, in all but one instance we strongly suspect the Yellow Boa (Epicrates subflavus) because of the "lack of evidence" (i.e. mammals and other birds will take eggs or leave gnawed remains; boidae snakes almost never take eggs and consume entire clutches early in the nestling period or individuals later in the cycle when nestlings are larger (see Marini and Melo 1998). Of the nest sites that fledged at least one chick (n=28), 13 (46%) experienced brood reduction. Discounting neonate mortality of the last-hatched chick, which was observed in nearly all instances of clutches of four, seven nestlings died of unknown causes late in the breeding cycle and one disappeared to predation.

The behavior of adults, particularly females, corroborates our belief that Jamaica's parrots have evolved in a system of intense selective pressures by native predators. During incubation, when probability of nest failure was quite low, females showed no hesitation to enter their nests following their return from a food-transfer bout (mean entry time following arrival = 2 + s.d. 1 mins; mean time off eggs during incubation was 16 + 6 mins, although some females left eggs unattended for nearly 2 hours during the warmest mid-day hours). As eggs began hatching (and probability of failure increased significantly), females became more vigilant (qualitatively by their behavior, included repeated aborted attempts to descend into the cavity) and took significantly longer to enter the cavity (mean = 9 + s.d. 11 mins). Additionally, in sharp contrast to other insular Amazona parrots which roost in the nest for the duration of the nestling period, Black-billed Parrot females ceased roosting at the same time they ceased continuous diurnal brooding. We are aware of one female that was disturbed in her nest during egg-laying/incubation at 0400; a burst of squawks was heard and the nest abandoned. We found two eggs buried 2 cm deep in the substrate and, again, we suspect disturbance by the nocturnal Yellow Boa. Several other persons
with extensive experience observing other reproductively-active Amazon parrot species agree that Jamaica's parrots are the most cautious at their nests (N.F.R. Snyder, R.Rattner pers. comm., S.Koenig pers. obs.).

Typical of other Amazon parrots, Black-billed Parrots are generalists in the types of tree hollows they adopt for use. Few physical characteristics of nesting cavities varied between successful and unsuccessful nests. The presence of but a few vines on the nest tree, however, was a significant predictor of nest failure (X2=9.87, df=2, P=0.007), a feature that has correlated positively with preferred foraging habitat of the Puerto Rican Boa (Epicrates monensis), congeneric to Jamaica's endemic boa (Chandler and Tolson 1990, J.W.Wunderle, Jr. pers. comm.). Additionally, nest failure varied as a function of the location of the nest tree (ANOVA, F=4.513, df=2,61, P=0.015). Nest sites located in badly degraded edge pasture (e.g., nest tree isolated from other trees; 39% failure) or in interior, minimally disturbed forest (40% failure) had significantly greater nesting success (Tukey HSD=0.795) than trees located in edge habitat of regenerating forest (75% failure; edge disturbed vs. edge regenerating: Tukey HSD=0.025; interior vs. edge regenerating: Tukey HSD=0.133). Presently, we lack information on Yellow Boa densities among these habitat types, but access to human-associated prey base (e.g., rats) may contribute to an artificially maintained population in edge habitat, although fatal encounters with humans in badly degraded edge habitats may limit snake populations. Modifications of individual nest sites to provide drier substrate conditions resulted in no significant improvements to nestling growth and fledging success. These results offer further support that properties of individual nesting cavities are not as important as the surrounding habitat (as it influences predator activities) in limiting nesting success of Black-billed Parrots. Perhaps more important, the reproductive rates in the regenerating edge habitat are insufficient to maintain a viable population of parrots. The species has a very high probability of going extinct should reproduction among all the habitat types fall below 75%, an event which is plausible should humans continue to press further into the interior, virgin forest and disturb the structural integrity of the forest.

Where native forest still exists, Jamaica's native parrot populations should be considered locally common. As obligate cavity nesters, the preservation and effective management of these remaining forests is critical to ensure viable populations. The overall annual nesting success of Black-billed Parrots is low for an insular parrot species, but quite acceptable for a long-lived species whose individuals have many opportunities to reproduce during their lifetime. These life history traits, however, may mask the long-term viability of a population, particularly as the presence of an adult population does not guarantee reproductive activities (e.g., source/sink dynamics (Rosenzweig 1985, Pulliam 1988, Danielson 1992)).

A closer examination of breeding performance along a habitat gradient reveals that regenerating forest in edge habitat, while having a quantitative abundance of tree hollows in which parrots may breed, qualitatively represents poor habitat because of the high predation rates. However, an alternate view is that this habitat appears optimal for predators, which may be equally, if not more, endangered than their prey. The evolutionary balance between predators and prey are most likely observed in the interior, minimally- disturbed forest of Cockpit Country. Increasing human activities in the interior (e.g., creation or widening of trails, removal of saplings for yam sticks -- which removes small trees and prevents regeneration of old-growth forest while also disturbing the canopy structure) are expected to have direct impacts on predator/prey dynamics. A major conservation goal should be to maximize the ratio of interior:edge habitat (i.e. single, large habitat of Cockpit Country, rather than allowing many trails/corridors which isolate individual hills) in order to maintain the functional ecosystem.

As a cautionary note: attempts in 1998 to accurately identify predators of parrot nestlings failed to achieve desired results. We attributed nearly all predation to Yellow Boas through a process of elimination, rather than direct evidence. The majority of parrot nest sites are nearly two meters deep, which precludes avian predators such as the Jamaican Crow (Corvus jamaicensis) from reaching nestlings. As stated previously, we never have found evidence of mammalian predation. Our strongest evidence occurred at a Yellow-billed nest site, where nestlings were observed to fledge. Upon our return to collect nest-tree measurements, we found a snake pellet with the legbands of one of the chicks and a shed snake skin; the following year this site was adopted for use by Yellow-billed Parrots and, shortly after eggs hatched, the nest contents disappeared but a snake skin was found at the entrance to the nest. We strongly recommend continued research:
(1) to positively identify predators through the use of video or photographic means;
(2) to develop effective predator deterrent traps should predation rates increase to unacceptable levels;
(3) implement a study of Yellow boas to assess densities along a habitat gradient, determine habitat preferences and monitor foraging activities. This latter recommendation is important, particularly as many other native birds, reptiles and mammals comprise the diet of this endangered, endemic snake.

REFERENCES We value your feedback and comments: