The draft Environmental Assessment for release of `alala to several sites on the Island of Hawai`i describes each prospective release site in some detail. As outlined below, each site poses its own unique set of challenges:
Pu`uwa`awa`a. The Pu`uwa`awa`a sanctuary has “few structural impediments” to management of threats to `alala. Predator trapping has previously been conducted and rehabilitation of native vegetation is a management priority,” the draft EA states. However, as with the `io, there are some intractable difficulties, especially with control of banana poka. This vine presents a threat to native forests, which it smothers with its dense foliage. However, the juicy banana poka fruit (closely related to passion fruit) is potentially an excellent source of food for the released `alala, especially in light of the absence of native plants that the `alala may have once used for food. “Control of this vine is a priority for management within the sanctuary,” the draft EA states, “creating a potential conflict between weed management and `alala recovery needs if the sanctuary were to be a release site.”
Fire is another issue, though not directly addressed in the draft EA. In 1995, wildfire raged through the eastern one-quarter of the sanctuary, “reducing its carrying capacity for native forest birds,” the draft EA notes. A proposed management plan for the sanctuary calls for “pulsed” cattle grazing to control alien grasses that fuel such fires. This, the draft EA says, “could interfere with predator trapping and affect revegetation rates and other activities.”
Also, the sanctuary adjoins a game management area. “Educational materials for hunters É would be needed to explain the release program and to help ensure that `alala would not be confused with game birds,” the document notes. Feral pigs are “abundant” in the sanctuary.
The sanctuary is also relatively small, capable of holding just four pairs of `alala. Should the birds fly into nearby areas (including ranch lands and a residential subdivision), they could be exposed “to mortality factors that cannot easily be controlled outside of state lands. These include predators, especially fleral cats and domestic cats associated with residences, and shooting,” the document says.
“Over time,” it goes on to say, “`alala released at the Sanctuary could establish nesting territories in suitable habitat elsewhere on Hualalai and in North Kona. However, the amount of suitable habitat is declining. Between 1976 and 1994, satellite images show that approximately 800 hectares (1980 acres, or 10 percent) of potential nesting habitat was converted to non-forest.”
Honomalino. This 2,900-acre state-owned tract was grazed until recently, areas adjoining it have been and continue to be logged for koa, all conditions of vegetation seem to be favorable for `io, feral pigs are abundant – to the point they’re interfering with a rare plant restoration program – and even feral cattle continue to be observed from time to time. All in all, it is difficult to understand why Honomalino was included among sites considered for re-establishment of `alala populations. About all the draft EA says on this point is that “the South Kona area around Honomalino held one of the last subpopulations of `alala. The potential release site apparently contained a single nesting pair of `alala in the late 1970s.”
Kapapala. This area of 3,850 acres in the district of Ka`u includes a tract, known as the Koa Management Area, which the state only recently withdrew from a ranching lease in order to “experiment with sustainable koa forestry methods.” `Alala were seen in the area as late as the 1950s.
“The reasons for the extirpation of `alala from Kapapala are obscure,” the draft EA notes. “Populations of other forest birds persist here, and no logging or commercial grazing has occurred in the forest reserves. Limiting factors É may be similar to those found on McCandless Ranch: predation, disease, and perhaps food limitationÉ. Limiting factors are likely still present, due to little forest recovery or active habitat management over the years.”
Should `alala be released here, the draft EA states, “management actions É would need to be planned and integrated with existing and proposed activities at the site, which include hunting, gathering of native plants, and sustainable koa logging.”
As at Pu`uwa`awa`a, “the total amount of potential nesting habitat in Ka`u is declining. Between 1976 and 1994, satellite images show that approximately 1,945 hectares (4,800 acres, or 17 percent) of potential nesting habitat was converted to non-forest.”
Kulani. The proposed release site here consists of just over 4,000 acres of land in private and state ownership. Although it has much good forest cover, the area has never been known to be inahbitated by `alala, even when the species was more abundant. As the draft EA states, “It is possible that the high rainfall downslope of the Volcano area, directly or indirectly, makes these forests unsuitable for `alala.” `Alala nests have never been seen in areas where more than 80 inches of rain falls annually. At Kulani, 93 percent of the area under consideration receives more than 80 inches. “Therefore,” the draft EA states, “if high rainfall limits `alala distribution,” the nesting habitat at Kulani may be far less than” suggested by tables within the document. Those tables indicate that more than 8,700 hectares (21,000 acres) of suitable `alala nesting habitat is available at Kulani and nearby areas.
Hakalau. About 2,500 acres lying within the Hakalau Forest National Wildlife Refuge is the last site being considered for possible `alala introduction. “Although single `alala were observed near Hakalau in 1971 and 1975,” the draft EA says, “observations of `alala far outside of their normal range were not uncommon in the 1970s. These observations were presumably of birds from collapsing populations searching for conspecifics” – i.e., other `alala.
The proposed release site is largely free of ungulates. About half is described as “invaded, but apparently suitable, nesting habitat.”
As with Kulani, high rainfall might be a deterrent to establishment of `alala. At Hakalau, annual rainfall is between 120 and 140 inches. Just 2 percent of the nesting habitat at Hakalau is under the 80-inch-a-year rainfall level that scientists believe `alala require. Again, the tables in the draft EA that show as much as 13,000 acres of suitable nesting habitat at Hakalau “may overestimate that which is actually suitable for released `alala,” the draft EA notes.
Status Quo
Whatever the outcome of the draft environmental process, the Fish and Wildlife Service will continue to release `alala at the McCandless Ranch site, the document says. This area includes about 2,250 acres of land owned by the state, private parties, and the federal government in the South Kona district of the Big Island. (The parcel owned by the state, known as the Waiea Tract, lies between the McCandless Ranch lands to the south and the Kona Forest Unit of the Hakalau Forest National Wildlife Refuge to the north.
The land has been grazed for decades, the draft EA notes, “but the low stocking densities used have allowed much of the understory vegetation to persist.” Feral pigs “are present in moderate numbers” and “are currently interfering with the predator trapping effort by disturbing the traps” for cats and mongoose, the draft EA states.
Game birds are numerous, including kalij pheasant, turkey, and California quail. “Both avian pox and avian malaria are present at high levels which appear to fluctuate seasonally,” the document notes.
“Essentially the entire current release area is nesting habitat, but only the lower one-third is classified as having a native-dominated understory,” according to the draft EA. “This area can potentially hold three to four pairs of `alala, approximately the number that occupied it in the late 1980s. A large area (28,600 acres) of nesting habitat is contiguous with the release site, and could hold 29 pairs of `alala if habitat quality were similar to the release site.”
The total area of nesting habitat in and around the McCandless Ranch area is estimated at 39,600 acres. This, the draft EA suggests, could accommodate up to 40 pairs of `alala. “However,” it goes on to note, “unless existing limiting factors are controlled through much of this area, it is questionable whether mortality rates could be reduced sufficiently for `alala to reestablish and persist over this range. Present habitat quality is suspected to be similar or worse than that which existed during the contraction and disappearance of the historical `alala population in Kona during the latter half of this century.”
— Patricia Tummons
Volume 10, Number 8 February 2000