Hess et al. • MOVEMENTS OF THE HAWAIIAN GOOSE 
479 
altitudinal migration may allow Nene to track 
availability of food resources not otherwise 
seasonally available (Black ct al. 1997), (2) 
migration may enhance survival during the non- 
breeding season by avoiding non-native predators 
in breeding areas. (3) Nene may be able to avoid 
multiple anthropogenic hazards by occupying 
undeveloped high-elevation areas during the non¬ 
breeding season, and (4) there may be elevation- 
related environmental factors such as thermal 
constraints or humidity that affect the physiology 
of developing ova. embryos, and goslings (Ravel¬ 
ing 1978. Cooper et al. 2005), We used visual 
observations of banded individuals and satellite 
telemetry to document seasonal movements of 
Haw'ai'i Island Nene that have re-established 
traditional movement patterns. Our objectives were 
to: (11 identify locations used and the exchange of 
geese among breeding and non-breeding areas, and 
(2) understand the magnitude of elevation change 
and potential differences among individuals and 
breeding subpopulations by modeling altitudinal 
migration patterns. 
METHODS 
Study Area.—Nene primarily moved among five 
study areas across Hawai’i Island (19 37' N, 155 
28' W): the Kahuku unit of Hawai i Volcanoes 
National Park (Kahuku: 2.000-2.700 m elevation): 
Hakalau Forest National Wildlife Refuge (Hakalau 
Forest NWR; 1.800 ni): KTpuka ‘Amahou Nene 
Sanctuary (2.000 m); Kulani Correctional Facility 
(Kulani: 1.580 m); and the Big Island Country 
Club golf course (BICC; 640 m) (Fig. I). Long¬ 
term mean annual precipitation was 975 mm at 
Kahuku, 2,479 mm at Hakalau Forest NWR. and 
567 mm at BICC. The predominant ground cover 
at lower elevation sites was non-native kikuyu 
grass ( Pennisetum clandesiinum). Higher-elevation 
shrublands were dominated by native species in¬ 
cluding pukiawe ( Le/necophylla tameiameia ) and 
’ohelo ( Vaccinium reticulation) with sparse ground 
cover of native hair grass ( Desc/tantpsia nubigena). 
Hakalau Forest National Wildlife Refuge and 
Kulani were formerly natural wet forest areas 
unsuitable for Nene, but had been convened to 
several species of non-native pasture grasses for 
grazing except for small areas of native shrubland. 
Breeding and molt occurred where Nene were 
re-established after 1991 and 1996 at BICC 
and Hakalau Forest NWR. respectively. The other 
locations were used during the non-breeding 
season. 
Observational Data. —Visual observations of 
numbered leg bands were compiled to ascertain 
the proportion of Nene that moved among study 
sites. Nene had been banded by the Hawai‘i 
Division of Forestry and Wildlife (DOFAW) at 
BICC and Hakalau Forest NWR as part of 
ongoing management efforts. We recorded all 
identifiably banded Nene front 2009 to 2011, and 
obtained comparable data from DOFAW during 
2007-2011. Observations were collected regularly 
at Hakalau Forest NWR during 2009-2010. BICC 
during 2007-2011. and opportunistically at Ku¬ 
lani. and at remote areas of Kahuku. and the 
vicinity of KTpuka ‘Ainahou during 2009-2011. 
Satellite Telemetry.—Two cohorts of Nene 
were outfitted with Platform Transmitter Terminal 
(PTT) units equipped with Global Positioning 
System (GPS) capability (Microwave Telemetry. 
Columbia, MD. USA). PTTs measured 57 X 30 X 
20 mm and were attached dorsally with a double- 
threaded backpack harness made of Teflon® 
ribbon (Bally Ribbon Mills. Bally. PA, USA). 
PTT units were fitted only on males to reduce 
potential interference w'ith breeding and because 
mates are generally monogamous and travel 
together throughout the year: eight of 10 males 
were paired with females al time of capture. 
Transmitter packages weighed s3% of each 
bird's body mass. Capture, handling, and trans¬ 
mitter attachment procedures were approved by 
the University of Hawaii IAC-UC Protocol 08-636. 
Candidate None for satellite telemetry must have 
been observed at Kahuku and must have nested 
either at Hakalau Forest NWR or BICC. 
The first cohort of five males in 2009 was fitted 
with 40-g battery-powered PTT units programmed 
to take global positioning system (GPS) locations 
at 0000 and 1000 hrs HST. The second cohort of 
six males in 2010 and 2011. including one from 
the 2009 cohort, was fitted with 45-g solar- 
powered PTTs for a total of 10 sludy subjects. 
Solar-powered PTTs were programmed to take 
GPS locations at 0000. 1000. 1400. and 1900 hrs. 
All PTT units uploaded data to satellites every 
3 days (CLS America Inc.. Upper Marlboro. MD, 
USA). We conducted stationary trials for PTTs 
prior to attachment and 95% of GPS locations 
were horizontally accurate ±15 in. 
Statistical Analyses.—Locution coordinates 
were matched with a 10-m digital elevation model 
of Hawai'i Island to document movement and 
elevation (ESRJ 1999). Data consisted of Nene 
identity and origin, elevation, and ordinal date. 
