USGS PIERC EXPLORING METHODOLOGIES FOR THE REINTRODUCTION OF AN ENDANGERED HAWAIIAN FINCH

EXPLORING METHODOLOGIES FOR THE REINTRODUCTION OF AN ENDANGERED HAWAIIAN FINCH

Elizabeth M. Gray, John David Semones, Paul C. Banko, Steve D. Dougill, Dan M. Goltz,
Luanne M. Johnson and Michael R. Wiley

Pacific Islands Ecosystem Research Center; USGS BRD, P.O. Box 44, Hawaii Nat'l Park, HI 96718

Figure 1: Historic and current ranges of palila on the Island of Hawai'i.

INTRODUCTION

The palila (Loxioides bailleui) is an endangered finch-billed honeycreeper endemic to the Island of Hawai'i. Currently, the palila occupies less than 5 percent of its historic range (Figure 1). The extant population is located in subalpine dry forest on the west slope of Mauna Kea volcano, with grater than 90% of individuals concentrated within 25 km2. Natural expansion outside this core area is limited by a fragmented, degraded landscape and presumed strong site tenacity. As a result, the population is highly vulnerable to extinction by a single catastrophic event such as fire. To decrease extinction risk, we conducted a series of three reintroductions to a site on the north slope of Mauna Kea, approximately 18 km from the core population, where palila have been extinct for over 25 years. This area was chosen after comparing 1) food availability, 2) forest composition, structure and regeneration, and 3) disturbance from military training and other human activity among four possible reintroduction sites within the historic range. During 1997 to 1998, we reintroduced palila to the north slope of Mauna Kea by moving birds from the primary population on the west slope. Additional reintroductions are planned.

OBJECTIVES

Our objectives in this study were to develop reintroduction techniques that:
limited transport stress and mortality,
reduced emigration from the release site,
were economical and feasible
established a self-sustaining breeding population.

METHODS

Capture and Transport

We captured palila in mist nets from Mauna Kea's west slope core population. We banded, measured, and weighed each bird, and we pulled a tail feather to obtain pulp for DNA sexing.

We determined age using plumage characteristics and bill coloration.

We determined sex using plumage differences in nape and confirmed the sexes of 31 palila using DNA analysis.

Palila were fitted interscapularly with Holohil System Ltd ® transmitters using SkinBond ® and Super Glue ®. The transmitters had an eight week battery life and weighed less than 5percent (1.9 g) of the bird's body weight.

We modified a small pet carrier with foam padding, two perches, and fresh food for use as a transport cage. We suspended the carrier inside a 4x4 vehicle using a plank attached by two bungee cords. We covered the carrier to calm the bird. A person in the back seat of the vehicle stabilized the carrier.

Birds were transported 18 km (2.5 to 3 hour drive) to Mauna Kea's north slope.

Birds were weighed again prior to release and released near a large tree which offered both shelter and food.
Palila were fitted interscapularly with transmitters

Palila were fitted interscapularly with transmitters

Monitoring

We tracked birds on foot using hand-held radio receivers. We recorded their locations and observed their activities multiple times per day. Roost locations were found when time permitted.

We mapped tracking locations using a GPS (plus 10 m) unit and ArcView 3.1.

Predator Control

We began trapping feral cats and mongooses and distributed diphacinone for rats to control introduced mammalian predators during reintroduction 2.

We trapped continuously from 10 September 1998 to 4 March 1999 for a total of 2628 trap nights.

RESULTS

AGE and SEX OF REINTRODUCED BIRDS

	Adult	Second Year	Hatch Year
Reintroduction 1	-	8 13 4 unknown	-
Reintroduction 2	3 3	4 7	5 2
Reintroduction 3	-	-	4 unknown

Table 1: Total number of palila moved in each reintroduction by age class and sex.

Reintroduction 1: We moved 25 second year birds (6 - 10 months of age) from February to March 1997 (Table 1).

Reintroduction 2: We moved 24 birds from October 1997 - March 1998: 6 adults, 11 second year birds, and 7 hatch year birds.

Reintroduction 3: We moved 4 hatch year birds from November - December 1998.

TRANSPORT STRESS

Percentage of body weight lost during transport, which we used as an indicator of transport stress, was low and fairly consistent during all three reintroductions (Figure 2).
The percentage of weight lost was significantly less than the weight loss incurred during an earlier reintroduction, in which 35 palila were moved in groups to a site on the east slope of Mauna Kea after being held overnight in large flight aviaries (Fancy et al. 1993; univariate ANOVA F3,81 = 4.183, P = 0.008).
In addition, the percentage of weight lost during transport did not differ among the three age classes of birds (univariate ANOVA F3,51 = 0.201, P = 0.82).

chart

Figure 2: Mean percentage of body weight lost during transport. Weight loss is compared to mean percentage weight lost in a previous reintroduction with different transport methods (Fancy et al. 1993). An * indicates a difference at the P greater than 0.05 level.

Capture site (West slope of Mauna Kea)

Reintroduction site (North slope of Mauna Kea)

FATE AFTER 8 WEEKS

Figure 3: Fate of birds at the reintroduction site during the first eight weeks following release.

Reintroduction 1:

During the first 8 weeks, 40% of individuals (n = 25) remained at the reintroduction site and 56% of individuals returned to the capture site. The fate of 4 percent of individuals is unknown due to premature transmitter detachment and battery failure (Figure 3).

Reintroduction 2:

In contrast, only 21 percent of individuals (n = 24) remained at the reintroduction site and 12.5% of individuals returned to the capture site. Many (46 percent) of the birds were depredated, presumably by introduced mammals and native raptors.

Reintroduction 3:

Two of four individuals remained at the reintroduction site and two returned to the capture site.

MONITORING

During the first 8 weeks after release, birds that returned to the capture site did not range farther than birds that remained at the reintroduction site (Figure 4). Most daily activity of reintroduced birds occurred within 1 kilometer of the release site, in an area with many large, dense trees.

Figure 4: Tracking locations of birds that remained at the reintroduction site versus returned to the capture site during the first eight weeks following release.

PREDATOR CONTROL

From September 1998 through March 1999, we removed 17 feral cats and 9 mongooses from the reintroduction site. Because we poisoned rats, the exact number killed is unknown.

LENGTH OF STAY AT RELEASE SITE

The total number of days that birds remained at the reintroduction site did not differ significantly among the three age classes (Figure 5; univariate ANOVA F2,53 = 2.350, P = 0.11, power = 0.45).
However, when we analyzed birds that stayed at the reintroduction site for > 8 weeks, we found that second year birds tended to stay for fewer days than adult or hatch year birds (univariate ANOVA F2,17 = 3.782, P = 0.05).
The total number of days that birds remained at the reintroduction site did not differ significantly for males versus females (Figure 6; t-test t = 0.257, df = 43, P = 0.80).
This pattern did not change when we limited analysis to birds that stayed at the reintroduction site for > 8 weeks.

total number of days by age class

Figure 5: The total number of days that palila remained at the reintroduction site by age class.

Figure 6: The total number of days that palila remained at the reintroduction site by sex.

CURRENT STATUS

An unbanded adult male appeared at the reintroduction site in December 1998. His origin is unknown.
As of early 1999, five banded birds remained at the reintroduction site.
To date, we have no evidence of breeding attempts at the reintroduction site, although last spring we observed a male and female courtship feeding and roosting together.

CONCLUSIONS

We have developed safe, effective and economically feasible methods for transporting palila.
Many palila return to their capture site within 8 weeks of relocation. More research is needed to understand whether factors such as differing habitat quality or site fidelity cause emigration rates to be high.
In some years, predation is the greatest source of mortality for reintroduced birds. Thus continued predator control is critical to reintroduction success.
Size of the daily activity range of reintroduced birds does not appear to influence whether palila remain at the reintroduction site or return to the capture site.
We found no difference between the length of time males versus females remained at the reintroduction site. However, adult and hatch year birds tended to remain at the reintroduction site longer than second year birds.

FUTURE DIRECTIONS

Our aims for future work are to:

Increase fidelity to the reintroduction site.
Decrease predation.
Promote breeding.

Specifically, we plan to:

move adult and hatch year birds together, to better determine which age class (if any) is least likely to emigrate from the reintroduction site.
investigate the development and role of philopatry by moving young (3 - 5 months of age) and older (6 - 10 months of age) hatch year birds.
move adults in early spring and late fall to determine whether there are seasonal differences in emigration rates.
continue predator control of introduced mammals.
move a greater number of individuals per reintroduction to determine whether emigration and a lack of breeding effort are influenced by infrequent social interactions between conspecifics at the relocation site.

ACKNOWLEDGMENTS: We would like to thank the U.S. Army Garrison, Hawai'i for funding the research. In addition, we would like to thank the United States Fish and Wildlife Service for their assistance, Division of Forestry and Wildlife for assistance and permission to work on Mauna Kea, The Peregrine Fund, and the many interns who aided with field work.

Palila Restoration

EXPLORING METHODOLOGIES FOR THE REINTRODUCTION OF AN ENDANGERED HAWAIIAN FINCH