The Hawaiian Crow (Corvus hawaiiensis), or 'Alala, once an inhabitant of large forested areas of the island of Hawai'i, is now found only in the wild in a relatively small area of the central Kona coast, specifically on the privately-owned McCandless Ranch. The decline of the 'Alala is part of a larger phenomenon of reduction and extinction of forest birds throughout Polynesia that has been associated with human colonization. Thus, its decline is not an isolated ecological event, but rather a symptom of underlying ecological problems. Of particular concern here is the ecology of montane forests on the western slopes of the island of Hawai'i, known as the Kona district. In other habitats along the Kona coast where the 'Alala has already disappeared, numerous endemic bird populations have also become extinct or are endangered and rapidly declining. The 'Alala is the most conspicuous member of this group and can be viewed as a classic indicator species.
Three bird species that were known only on the Kona slopes are already extinct: the Greater Koa Finch (Rhodacanthis palmeri), the Lesser Koa Finch (R. Flaviceps), and the Kona Grosbeak (Chloridops kona). They were seed-eating birds and fed primarily on koa and other seeds, on hard, dry fruit and seeds of the naio (Mycoporum sandwicense), and on lepidopteran larvae (Scott et al., 1986). All three were extinct by around 1900. It is reasonable to assume that the 'Alala's precipitous decline has occurred because the problems that caused the prior extinctions have not gone away. The systemic problems, and not just the decline of the 'Alala, must become the focus of scientific and conservation attention. As long as the 'Alala exists in the wild, it can help researchers to identify facets of the systemic problems at the ecosystem level, and perhaps provide insight on how managers can remedy them and prevent additional extinctions on the Kona slopes.
The Wild Population of 'Alala
The 'Alala is an omnivorous, but primarily fruit-eating, forest-inhabiting corvid. Both in its reliance on fruit and in its restriction to a forest habitat, the 'Alala differs from the widespread and familiar crows of the continents. The 'Alala is far more specialized.
As of April 1992, it is known that at least 11 'Alala exist in the wild in at least three and possibly as many as five breeding territories on the McCandless Ranch (J. Engbring, pers. comm., 1992). In spite of the precipitous decline elsewhere, numbers of 'Alala observed on the McCandless Ranch appear not to have changed drastically since 1976. The existence of other small population on adjacent lands cannot be ruled out.
The 'Alala used to be found in relatively low densities from the Ka'u District in southeastern Hawai'i to North Kona, north of Hualalai. Although extensive field studies have been conducted in those areas, many aspects of the natural history and behavior of the 'Alala are still poorly understood. Such information remains critical to developing and implementing appropriate preservation efforts. More information about the habitat is also necessary to develop efforts to preserve other native species.
A single cause for the decline in the wild 'Alala population cannot be identified; however, three categories of factors have been directly and indirectly implicated by previous studies: habitat and food, exotic predators, and exotic diseases and parasites. Agricultural development by the Polynesians destroyed most of the original dry lowland forests, restricting the 'Alala to largely wet, open, mid-elevation 'ohi'a (Metrosideros polymorpha) and koa (Acacia koa) forests. In the post-European settlement period, extensive ranching, logging, and the practice of allowing feral ungulates—primarily cattle (Bos taurus), but also pigs (Sus scrofa), sheep (Ovis aries), and goats (Capra hircus)—to roam in the forests have degraded the forests, reducing the native understory of fruit-bearing trees and shrubs. Even the current habitat of the remaining 'Alala is not necessarily prime habitat for the species. The extent of collecting and killing by humans is difficult to know, but it has probably been important, even in recent decades, and might have contributed to the high adult mortality of recent years off the McCandless Ranch. Introduced black (Rattus rattus) and Polynesian roof (R. exulans) rats prey on eggs and nestlings, and introduced mongooses (Herpestes auropunctatus) and cats (Felis catus) undoubtedly prey on fledglings on the ground. Two introduced diseases that are widespread among native Hawaiian birds have been shown to affect the crow: avian malaria and avian pox.
The committee's analysis of data obtained from census efforts during the 1970s and 1980s on a small sample of banded 'Alala reveals that during that period of precipitous decline clutch size of the 'Alala was somewhat smaller than temperate Corvus species, but sample sizes are small and comparative data for other tropical insular species are lacking. On the McCandless Ranch, the 'Alala continued to produce fledglings at rates somewhat lower than those of temperate corvids, and fledging rates might even have been slightly lower elsewhere on the Kona coast. Juvenile (0–1 year) survival in the Kona District was comparable with that of other corvids that are not endangered. Except on the McCandless Ranch in the last 14 years, however, death rates of adult 'Alala in the wild were inordinately high. We do not know why adult survivorship has been lower than would be expected for a Corvus species or why it has been higher on the McCandless Ranch than at other sites.
What follows is a summary of the committee's major findings and recommendations. Each is examined in further detail in Chapter 7.
The federal Alala Recovery Plan (Burr et al., 1982) and the state Alala Restoration Plan (Burr, 1984) are admirable documents that contain concise summaries of the history and status of the species and sound recommendations for recovery. Each places priority on the protection and restoration of native habitat, the study of disease and predator control, and recommends the management of the captive and wild populations. The federal plan designates critical habitat that needs protection, and the state plan sets priorities for how that land should be incorporated into a conservation scheme. In the decade since the federal plan was issued, however, it has not been implemented, and the 'Alala has declined further.
A reliable estimate of the number of 'Alala that would constitute a minimum viable population for long-term survival of the species is not now possible. Given the small size of the current population, the species will be in danger of extinction for a long time. We conclude that there is a high probability that the current population in the vicinity of the McCandless Ranch will become extinct as a result of chance events in 1–2 decades unless its numbers and geographic range are increased. The committee is unanimous in its opinion that only an active management program can prevent the wild population from becoming extinct.
We do not recommend that all the birds should be brought into captivity. A viable population of 'Alala in the wild should be maintained and increased. To avoid extinction, the sizes of both the wild and captive populations must be increased to provide demographic and genetic security. When populations are as small as these, demographic accidents and random environmental disturbances are likely to cause extinction. Considering the small size of the remnant populations, the committee recommends joint management of the wild and captive populations as a single unit. Joint management will require that the identity of all existing birds be known. The possibility that additional 'Alala survive on both public and private lands in Hualalai, Honaunau, and the Ka'u District needs to be thoroughly investigated by the U.S. Fish and Wildlife Service and the state of Hawai'i, because additional birds would be crucial to recovery of the species. Any additional birds would both increase the existing gene pool and decrease the probability of demographic accidents.
The goal of joint management of the captive and wild populations is to increase the density and distribution of the population as rapidly as possible. To accomplish the joint management of captive and wild populations, a new recovery team or advisory working group for the 'Alala should be established that includes state and federal biologists; other professional biologists who are experts in avian ecology, captive propagation, reintroduction, long-term population biology of birds, and population genetics; an avian veterinary pathologist; an aviculturist; and a representative of the private sector, preferably a private landowner or land manager. This combination of experts will provide the knowledge and skills necessary for joint management of the wild and captive populations. The recovery team must work together continually to plan, coordinate, and implement all aspects of the recovery plan. A long-term, well-funded arrangement must be established.
For recovery and sustainability of the 'Alala, as for other endangered species in Hawai'i and elsewhere, habitat maintenance and restoration are essential. Unless the causative factors of decline are identified and corrected, preservation efforts for the 'Alala will be compromised. Through habitat preservation and holistic management of cattle ranches, and control of predators, some potential causative factors can be mitigated.
The committee strongly urges the state of Hawai'i and the U.S. Fish and Wildlife Service to establish at least one major forest preserve along the Kona coast. In addition, cattle ranches should be managed in ways that provide critical habitat for native flora and fauna. The habitat of the wild population of 'Alala should be managed aggressively to control predators—including mongooses and rats—and the impacts of exotic plants and feral ungulates.
Techniques must be developed for managing the wild population without risk of injury to the birds. Increasing their numbers in the wild will require knowing what they need to survive, remain healthy and reproduce successfully. Most of the information needed to support the management of the 'Alala as its numbers increase can be gathered noninvasively, with minimal disturbance of the wild birds. Some basic biological facts about their social behavior are still unknown. Most urgently needed are data on habitat requirements, variations in food resources available and required through the seasons, foraging behavior, physiology and disease, social behavior, and demography. Those data will provide specific direction for management of the property that supports the final wild population as 'Alala numbers increase and of additional forest preserves after reintroduction or recolonization. Short-term measures that can be started immediately include improved forest management, control of predators, supplementation of the 'Alala's food supply, study of disease in introduced species of birds that are living in the same habitat, and a public-education program.
Long-term measures that require additional research, planning, and the cooperation and coordination by a recovery team include development of a safe and effective vaccination program for avian pox and control of other diseases and reintroduction of young hatched in captivity. The latter actions will require careful analysis of methods used for the reintroduction and translocation of other endangered species.
The Captive Population of 'Alala
Through research and experimentation biologists have developed manipulative techniques that have been highly successful in the preservation of endangered species. In the case of the Peregrine Falcon (Falco peregrinus), for example, captive breeding and reintroduction enabled the restoration of breeding populations in large portions of its former range at a rate far exceeding that which would have occurred through natural colonization. Those techniques have usually been applied to complement conventional field research and management efforts to preserve habitat and restore threatened populations.
Owing to the small number of 'Alala now remaining, complete recovery might seem hopeless; however, similar cases around the world that have used intervention measures have had striking success and offer hope that coordinated recovery efforts can be successful. To date, 39 species of corvids, including 9 Corvus species, have been bred successfully in captivity.
The management and husbandry of the captive 'Alala at Olinda captive breeding facility are inadequate, and there is room for modification and improvement of existing methods and procedures. The Olinda facility should be expanded. Management of the entire population at one site, however, is risky. The committee concludes that a single propagation facility is not adequate. A second facility should be built that provides an optimal environment for successful breeding of the 'Alala. An optimal facility would have additional crow enclosures that enable the safe capture and manipulation of the captive birds and incorporate the designs of experienced aviculturists from other captive-propagation programs. The general characteristics of an optimal facility (or facilities) would include institutionalization of routine and emergency veterinary care, adequate on-site veterinary facilities, a consistent pathology program, and an up-to-date library with access to journals and information at zoos on the mainland. Continual review of the avicultural literature and communication with colleagues are essential for developing or applying new techniques and procedures. A second facility should be established on the Hawaiian islands to minimize the potential for loss of the captive population to a disease outbreak or other catastrophe.
Other actions require extensive planning, cooperation, and coordination. For example, state and federal agencies should seek advice from outside experts about the design of an optimum captive habitat for the 'Alala. Consideration should be given to developing a more "natural" captive environment, constructing lower and wider breeding enclosures to facilitate capture while minimizing the risk of injury or trauma, providing breeding enclosures with multiple compartments to allow for the quick isolation and capture of males, provisioning each breeding enclosure with multiple nest baskets to stimulate nesting, and increasing the size of juvenile enclosures to promote improved flight conditioning. Staff training will be required, and protocols for standard and nonstandard procedures will have to be developed, circulated widely for peer review, and then implemented.
Husbandry in the captive-breeding program must be improved. Both short- and long-term actions are recommended. Some actions that can be instituted immediately are re-mating birds that have been paired unsuccessfully for more than 2 years, modifying enclosures to facilitate capture of and removal of males after the first egg of each clutch is laid, allowing females to incubate eggs 5–7 days before removal for artificial incubation, monitoring daily food intakes to assess dietary components and overall nutrition, and feeding the birds a more frugivorous, low-acid, low-iron diet.
Well-trained personnel are also essential. To remedy current shortcomings, a full-time director should be hired to serve as both a program administrator and curator. The person
should be knowledgeable about aviculture and ornithology and be up to date on avicultural techniques, and would be responsible for the overall training of keeper staff, time management, facilities administration, public relations, and fund-raising. A full-time on-site avian veterinarian, a full-time on-site aviculturist, and at least two additional animal keepers are also needed. Equally important is the establishment of a long-term avicultural training program for staff.
In addition to the establishment of more than one captive-breeding facility, longer-term actions include the development of techniques for semen collection and artificial insemination, expansion of the captive flock until there are at least 40 productive pairs, and development of chick-rearing protocols that maximize proper socialization and independence. Those actions will require additional funding, training, planning, cooperation and coordination.
Modern equipment is needed if the captive-breeding program is to be successful. The program should have a full veterinary laboratory and clinic (equivalent to the facilities of a small-animal practice) that includes quarantime areas, an x-ray machine, updated video-monitoring equipment, a library, and a hatching room and brooding facility. It also needs access to a pathology laboratory on the Hawaiian islands, protocols for pathological analyses that can be performed on site, and long-term arrangements with a board-certified pathologist or pathology center for pathological studies.
Although the historical decline of the 'Alala must have been accompanied by a loss of genetic variation, it is not clear whether the extant wild population is suffering from inbreeding depression caused by the loss of genetic variation. Recent DNA analysis shows the captive flock to be moderately to highly inbred. The 'Alala population on the McCandless Ranch is so close to the wild sources of some of the captive birds, however, the addition of wild adult birds from the ranch to the captive population for genetic reasons alone would be expected to provide no more than a very minor measure of new genetic variation to the captive population, and their removal from the wild would adversely impact on the extant wild population's potential productivity. Any potential beneficial genetic augmentation of the captive flock should be accomplished by removing eggs from the wild and hatching them in captivity.
Even extant genetic variation can be retained only if the population is increased rapidly. New genetic studies of the wild 'Alala are not recommended over the short term because the numbers are so small and the amount of information gained would not affect the principal recovery or habitat management actions that need to be implemented now. The capture of additional adult birds for captive breeding for genetic reasons alone should have a very low priority, because the potential genetic benefits would probably be minimal, and because wild-caught adults would have a reduced likelihood of breeding. Similarly, the release of captive birds on the McCandless Ranch solely for the purpose of augmenting genetic variation is not
advised. Any releases must be part of a full-scale management plan that has demographic augmentation as its primary goal.
Options for Management of the 'Alala
The committee's report suggests several options for the management of the 'Alala. We give the highest priority to the removal of first-clutch eggs from the wild population beginning in the 1993 breeding season. Egg removal ("egg-pulling," "double clutching," "multiple clutching") has become a successful method for augmenting the natural reproductive output of wild birds, and has been used successfully with other endangered species. It relies on the capability of most female birds to renest after clutch removal or to continue laying eggs beyond the normal clutch size if eggs are removed in sequence as laid. On the basis of what is known about other corvids and birds generally, the failure to renest after removal of the first clutch will probably not be more than an infrequent occurrence or a rare peculiarity of particular females.
We give the egg-removal option a high priority because it allows for simultaneous augmentation of the captive and wild populations without removing adult birds from the wild or seriously compromising the wild population's social organization or productivity. The use of egg removal and later reintroduction of young into the wild is obviously a long-term strategy and requires an active recovery team and optimum incubation and chick-rearing facilities on or near the McCandless Ranch. Results should be evaluated annually so as to make recommendations about what to do with the crows that are produced in captivity.
The great advantage of egg removal is that it provides a way to take substantial numbers of eggs from the wild for hatching and rearing of young without depriving the wild birds of opportunity for natural reproductive output in the wild. The combination of natural and artificial hatching and rearing yields many more young than could be produced by the unaided wild population. The young reared in captivity can be used in several ways: they can be released back into the wild, they can be retained in captivity to augment the captive-breeding stock, and they can be used to learn more about the biology of the species.
Optimum conditions for artificial incubation have not yet been worked out for the 'Alala, but this deficiency can be partly offset by ensuring that the wild eggs receive some natural incubation (about 5–7 days) before they are removed. Efficiency and success would be increased by locating the incubation and chick-rearing facilities near the wild population, i.e., on or near the McCandless Ranch. Also, what is learned from suggested modifications of the current method of incubation at Olinda should improve the hatchability of both wild and captive-produced eggs. Infertile eggs, dead embryos, and dead chicks that are collected and examined can provide important information for understanding the effects of inbreeding on reproduction and the role of artificial incubation and nutritional influences on embryo and chick development.
Recovery programs around the world have yielded information relevant to 'Alala preservation efforts. For example, wild populations of many threatened species have declined to extremely low numbers, yet have responded positively to well-conceived and carefully implemented programs. Many seemingly hopeless cases turn out to be salvageable when imaginative research and conservation programs are implemented in conjunction with concerted efforts to preserve and restore natural habitat.
Intensive in situ manipulative techniques can often lead to more effective and less-expensive augmentation of wild populations than ex situ, or captive-breeding programs, although both approaches might be required initially until the number of wild breeders is increased sufficiently. Many recovery programs have enhanced the fecundity or survivorship of wild birds considerably through the judicious application of predator control, supplemental feeding, and multiple clutching.
There are now several cases in which the geographic range of an endangered species that had been reduced to single, relict population has been expanded geographically through deliberate releases into suitable habitat. Some species, such as the Mauritius Kestrel (Falco punctatus), have even been successfully induced to colonize habitats that differ from their original ones. Most programs have deliberately focussed on re-establishing multiple populations to minimize the possibility of extinction as a result of local catastrophic events.
Because suitable habitat is necessary for the re-establishment of wild, self-sustaining populations, habitat preservation or restoration must have high priority in every recovery program. In the case of the 'Alala, that is especially true—essentially no native habitat remains in its pristine condition. Restoration and protection of forest preserves will have the added benefit of rescuing numerous other endangered plants and animals from similarly precarious situations.
Success in preservation of endangered species is possible only when the right combination of financial support, people, and techniques can be sustained. Although the committee recognizes that its recommendations will require the commitment of additional financial resources, it did not determine what those costs will be. But it is known that a successful recovery program will require funding continuity, effective administrative organization and commitment, creative and dedicated staffing, and effective communication and cooperation among the concerned parties. Most of those programmatic requirements are obvious, but the critical importance of coordination and cooperation needs to be emphasized. Recovery programs are never conducted in a vacuum, and they typically affect, directly or indirectly, diverse interest groups represented by government agencies, academic institutions, conservation organizations, and private land-holders. Without the support and cooperation of all those parties, decision-making will usually be determined in the political rather than the biological arena, and recovery actions may be delayed or thwarted.