4
Major Findings and Criteria for Recovery of the Aga
To develop the scientific criteria for a recovery plan for the aga, the committee reviewed and analyzed the available scientific information, was briefed by biologists from the US Department of the Interior Fish and Wildlife Service (FWS) and National Biological Service (NBS), the Guam Department of Agriculture Division of Aquatic and Wildlife Resources (DAWR), and the Commonwealth of the Northern Mariana Islands (CNMI) Department of Lands and Natural Resources Division of Fish and Wildlife (DFW), and spoke with others with relevant expertise, including US Department of Agriculture Animal Damage Control (ADC) personnel. Based upon the information detailed in the preceding chapters, this chapter summarizes the committee's major findings on the current state of knowledge of the aga's plight, and enumerates the basic criteria for prioritizing our recommended recovery actions. Chapter 5 evaluates various recovery options and compares the aga's situation with that of other corvids, including the 'alala in Hawaii. Finally, chapter 6 offers 11 specific recommendations for recovery actions to be taken in the next three years.
SUMMARY OF MAJOR FINDINGS
-
The present population of the aga on Guam is not reproducing and is not viable. The species was once distributed throughout the island of Guam, but is now found only in limited forested areas of Andersen Air Force Base at the northern end of the island. The number of adult birds has declined from an estimated 50 in 1992–1993 to 20 in 1996. There are no known breeding pairs at present and only 7 nestlings are known to have fledged on Guam since 1986. Available
-
information strongly suggests that males outnumber females in the current population, and that the age structure of the population is skewed towards older birds.
-
The only viable population of the aga is on the island of Rota and it appears to have declined by at least 50% during the last decade. The most recent estimate of 592 aga on Rota in 1995 was made with a survey method (variable circular plots) that the committee believes is likely to overestimate birds like crows (Ralph and others 1993). Even so, comparisons using the same transects and the same detection methods lead to estimates of either 50% or 70% declines since 1982, depending on how the data are analyzed. These rates of decline are substantially higher than estimates of rates of loss of forested habitat during the same period.
-
Predation by the brown tree snake is the most likely primary cause of the decline of the aga on Guam. Coincident with the decline of the aga on Guam has been an explosion in the population of the introduced brown tree snake. This snake is also the most probable cause of the extirpation of several other native forest birds on Guam. The fact that both the decline of the aga and other forest birds on Rota have not been more severe, and that the brown tree snake has not been detected on Rota, support the hypothesis that the brown tree snake is the primary cause of decline of the aga on Guam. The only management tool that has been used to mitigate the effects of brown tree snakes on aga has been the placement of electric barriers on nest trees. Additional studies are needed to assess the effectiveness of canopy breaks in stopping migration into nest trees and the effectiveness of trapping snakes that are present in nest trees when the barriers are installed. Snakes have been successfully eradicated from exclosures as large as 1 ha and trapping might be effective at reducing snake numbers in areas up to 7 ha. Neither intensive trapping of snakes nor use of exclosures has been tested on a scale of tens of hectares in predominantly forested habitats, the scale that will probably be required to protect the aga.
-
Inadequate precautions are being taken to prevent the spread of the snake to other islands. Although the brown tree snake has not been found on Rota in natural habitats, individual brown tree snakes have been found on at least 10 other islands, including Oahu in Hawaii, the US mainland, and elsewhere; a population might already be established on Saipan. With many forested areas on Guam still having densities of over 30 snakes/ha and the urgent need to prevent snakes from being carried in air or ship cargo to other islands, control of the brown tree snake must be given the highest priority.
The ADC program has attempted to reduce the dispersal of snakes in air and ship traffic from Guam since 1993. This program depends to a great extent on visual searches with hand capture, trapping, and detector dog and handler teams and to a smaller extent on habitat and prey-base management near transportation
-
facilities on Guam. The large numbers of snakes removed from within, around, and in moderate proximity to the transportation network make evident the risk of dispersing snakes to other islands and the need for additional snake control on Guam and other sites where they are likely to arrive. The number of snakes caught in 1995 is large (6,134), but verification of control efficacy, increased efficiency, and more comprehensive coverage of the transportation system are needed to adequately safeguard other geographic areas. Adequate justification exists to increase ongoing efforts in scope and intensity, and to simultaneously expand research to develop better, more effective technologies applicable to the large scale of the problem and the need for a multilayered safety net to prevent snake dispersal and colonization. The NBS program has emphasized the development of barriers to dispersal to augment other control techniques, arranged as exclosures to create snake-free areas. Effective control will probably require an integrated program that employs several techniques simultaneously. Neither DAWR nor DFW has aggressive management programs to control the brown tree snake.
-
Although based on limited data, estimates of genetic differences between the Guam and Rota populations indicate that differences do exist but are very small. Unique genetic diversity might reside in the Guam population (Tarr and Fleischer 1996), but any management decision based on genetic considerations should be considered in conjunction with the demographic goals of maximizing productivity and minimizing mortality rates (Lande 1988).
-
Aga nests fail at very high rates on Guam, compared with rates on Rota and rates in other corvids. Aga eggs or chicks in unprotected nests were almost always eaten by snakes or lost to other causes. Efforts by DAWR to install electric barriers around nest trees and to isolate nest trees from the adjacent canopy has increased clutch sizes and reduced losses of nestlings to brown tree snake predation, but have not increased the number of fledglings. Many nests are apparently abandoned either before eggs are laid or just afterwards, or the eggs are being eaten by brown tree snakes before researchers can detect them.
-
Eggs that have been taken from aga nests for artificial incubation have shown a high rate of developmental failure. Either many eggs are infertile or the embryos die in early developmental stages. That could be because the parent birds are getting too old to breed normally, because incubating birds are harassed by brown tree snakes, or because of other factors. The current state of knowledge is insufficient to distinguish between senescence and other causes of developmental failure.
-
Little is known about the demography, ecology, and behavior of the aga on either Rota or Guam. Field work on Rota has been limited by lack of permanent
-
staff supported either by DFW or by FWS. On Guam, adult mortality has not been directly studied, but territory turnover rates suggest that aga are disappearing more frequently than expected relative to other corvids. Because previous work has not involved studies of marked birds, either banded or tagged with radios, little is known about survivorship rates, the specific reactions of crows to the brown tree snake, why the birds on Guam seem unable to complete nest construction, and why the proportion of eggs that are viable is so low. Too few demographic data are available for viability analyses to be useful at this time. Studies of radio-tagged birds on both islands could provide information on causes of mortality and environmental stresses, and could yield needed demographic data. Comparisons between Rota and Guam could indicate similarities and differences between the two populations that could lead to insights about limiting factors. Such work should be done in conjunction with an assessment of the size and stability of the Rota population. Every effort must be made to ensure that Rota remains snake-free and that adequate habitat for the crow population is protected.
-
A captive population of 6 male and 6 female aga exists on Guam and the mainland. Of 10 adult crows taken from Rota by the Mariana Archipelago Rescue and Survey (MARS) program for captive breeding between 1993 and 1995, nine survive. Although 8 birds have been held as pairs (some with inadequate space for reproduction), only 1 pair produced 1 offspring. The current ten MARs birds are held in mainland zoo facilities. Two eggs produced in the wild on Guam in 1995–1996 hatched in captivity, and the resulting juveniles are being held in Guam.
-
A limited amount of aga habitat is protected on Guam and Rota. Habitat protection on Guam consists of about 1,600 forested hectares owned and managed by the government of Guam and 9,000 forested hectares recently incorporated into the Guam National Wildlife Refuge, mostly on Andersen Air Force Base. On Rota, most of the land is privately owned, and pressure for economic development is increasing. In an attempt to ensure that adequate forest habitat will be protected as other areas undergo development, a habitat-conservation planning process has begun on Rota. While the amount of habitat set aside on Rota and Guam could probably support a viable population of aga, the fact that the established wildlife conservation areas on Rota do not currently designate any areas known to be occupied by crows argues for protection of additional forest habitat. Forest habitat-protection and conservation-planning efforts on Rota should be accelerated.
-
Despite substantial efforts by many people, the overall level of cooperation and coordination among agencies and individuals involved in recovering the aga needs improvement. Data collected during management activities go
-
unanalyzed for years and have not been made accessible to appropriate groups or individuals. Off-island expertise has not been well integrated into recovery efforts. A coordinated recovery program is needed for the aga with a planned system of communication among participants and regular review by experts outside the program. Recovery of the aga is a complex issue that will require major activities and an organizational support system that is not now in place.
CRITERIA FOR PRIORITIZING AGA RECOVERY ACTIONS
In this section, the committee outlines its criteria for recovery of the aga. These criteria were used by the committee in evaluating the potential management options enumerated in chapter 5 and in formulating the specific recommendations contained in chapter 6. The committee believes that the following criteria are essential to develop a comprehensive recovery program for the aga.
-
Recovery of the Guam population is unlikely unless the brown tree snake can be controlled over multiple crow territories or tens to hundreds of hectares of contiguous forested habitat.
-
The most secure populations of an endangered species will serve as the source of individuals for long-term recovery. These populations tend to be the largest, the fastest growing, or the ones that are least affected by the factors that cause the species to decline. The core population of the aga is the one on Rota. Its security must have high priority.
-
A single population, no matter how large, is not immune to deterministic or chance events that can cause extinction. Each additional population decreases the likelihood that a single catastrophic event could extirpate a species. The aga is especially at risk because both Guam and Rota are subject to typhoons on the average of once every two years. The Guam population in its current, precarious state does not constitute a viable second population, so it cannot be considered an adequate hedge against the loss of the Rota population. This highlights the importance of restoring the Guam population, but it also suggests that the feasibility of establishing a population on another island should be considered.
-
Captive breeding of animals for release in the wild is slow, expensive, rarely results in successful reintroduction, and is not a good long-term conservation strategy (Snyder and others 1996). Captive propagation could be considered if the Rota population suffered a catastrophic decline, but captive breeding might divert attention away from needed field efforts.
-
Consideration of genetic variation within and between populations should be incorporated into population management actions whenever possible but not to the detriment of demographic goals. To the extent possible, within-population genetic variability in the Guam population should be maintained, but not at the expense of increasing the growth in numbers and populations of wild crows.
-
Management actions for the aga should be guided by research designed to determine the causes of decline and reverse the factors that limit the wild population. A coordinated management and research effort that facilitates the collection of data is needed to compare the ecology of aga populations of Guam and Rota, and to identify the limiting factors within each population. For example, such an effort would help to determine whether senescence or some other causes are responsible for the low viability of eggs on Guam. To the extent possible, key questions should be answered by using principles of experimental design: having control groups, using replication, and randomly assigning experimental groups to treatments.
-
Recovery programs should be part of overall efforts to conserve biological diversity and restore ecosystems. Endangered species recovery programs can promote the conservation of many species simultaneously if they are based on restoration of ecosystems. Because the brown tree snake has disrupted the entire forest ecosystem of Guam, its control in aga habitat could benefit other vertebrates. For example, areas in which snake populations were reduced by means of exclosures could be used as release sites for other endangered species, such as the Guam rail or the Micronesian kingfisher.