•Hanalei NWR was established in 1972 to provide recovery habitat for endangered species.
•Hanalei NWR supports five species of endangered Hawaiian waterbirds with statewide daytime counts <3,000 (USFWS 2004, USFWS 2011) as well as migratory waterfowl and shorebirds.
•The five endangered waterbirds include the ‘alae ke‘oke‘o (Hawaiian coot; Fulica alai), ‘alae ‘ula (Hawaiian gallinule; Gallinula chloropus sandvicensis), ae‘o (Hawaiian stilt; Himantopus mexicanus knudseni), koloa maoli (Hawaiian duck; Anas wyvilliana), and nēnē (Hawaiian goose; Branta sandvicensis).
•Hanalei NWR is essential to the continued conservation of these species, and supports the largest breeding populations of ‘alae ‘ula and koloa maoli in the Hawaiian island chain.
•Two primary threats to Hanalei NWR waterbirds are avian botulism type C and introduced predators such as feral cats (Felis catus).
•Avian botulism type C is a disease caused by a deadly neurotoxin produced by the anaerobic soil bacterium Clostridium botulinum. It is food poisoning. Birds usually die of respiratory failure or drowning (USGS 2015).
•From 2014-2018, 28% (756 of 2,868) of all sick and dead birds documented on Hanalei NWR were classified as botulism; on average 151 birds annually.
•Feral cats were introduced to the Hawaiian islands by European ships in the late 1700s (King 1984). Introduction of feral cats has had detrimental effects on native Hawaiian fauna through depredation and spread of disease (Underwood et al. 2014).
•From 2014-2018, 14% (390 of 2,868) of all sick and dead birds documented on Hanalei NWR were classified as depredation. Within that percentage, 64% (251 of 390) were classified as cat kills.
•In this case study we explore the relationships of habitat features on the levels of depredation on the refuge, and evaluate if carcasses can be used as an index of predator occurrence.
Study Sites and Time Frame
•This case study includes the year 2014 through the year 2018.
•Hanalei NWR is located on the north shore of Kaua‘i, Hawai‘i.
•Hanalei NWR consists of 371 ha (916 ac). Of that area, 64.7 ha (159.8 ac) are cultivated taro (kalo, Colocasia esculenta) and 49.3 ha (121.8 ac) are managed wetlands.
Botulism and Depredation Finds
•Sick and dead birds are collected in Hanalei NWR during avian botulism surveillance and incidentally during other activities by staff, volunteers, and taro farmers.
•Trained surveyors walk or drive along the open dikes surrounding fields, proceeding until all sides and interiors of fields are surveyed in search of intoxicated birds. During botulism outbreaks surveyors walk transects within fields.
•All major sources of decomposing protein are collected in order to prevent the spread of botulism.
•Depredations of birds are recorded when incidentally found during botulism surveys. Since surveys are designed to detect botulism intoxicated birds, these numbers are biased low.
•All carcasses are examined and classified as botulism, depredation, or other using field signs (e.g., cat hair in dead bird claws, gnawed bones, shearing and cutting of flesh and feathers).
•In the study time frame, there have been about 3,800 botulism surveys, excluding incidental finds.
•Feral cat management is conducted year-round, but primarily from November to July due to staff constraints.
•Primarily medium-sized, tomahawk-style live-traps are utilized.
•Main traps are set along the perimeter and major travel lanes and generally run 4 nights per week every other week. Response traps are set in response to cat detections, and run 1 to 8 days.
•All live-traps are set in the late afternoon and checked at least every 24 hours.
•Feral cats are humanely dispatched and kittens, adoptable cats, and domestic cats are turned in to the Kaua‘i Humane Society.
•Data collected include sex (male, female, and unknown) and age (kitten, young adult, adult, and unknown).
•Trap intervals (effective trap nights) are used to calculate effort; trap interval is calculated by (number of traps * trap nights) - (sprung traps * 0.5) (Beauvais and Buskirk 1999).
•Analyses were conducted using Excel and Google Earth Pro.
· The three most frequently collected species with depredation sign are endangered ‘alae ‘ula, ‘alae ke‘oke‘o, and koloa maoli, respectively, accounting for 76%. It is not clear why the ‘alae ‘ula account for 40% of depredations, but we suspect it is related to their regular use of dikes, which also serve as predator corridors.
· Male cats are caught more frequently than female cats. Both sexes are caught most frequently from March to June.
· Male cats are captured most frequently from March to May. This may be due to them dispersing in search of females and new territories.
· Female cats are captured most frequently from April to June. This may be due to them hunting to support their litters.
· Adult cats are caught more frequently than any other age class.
· Peak catch for adult cats occurs in May and June. This may be due to an inherent bias as trapping efforts are highest in April to June, but could also be related to cat breeding seasons.
·Young adult cat-catch numbers peak in April, even exceeding adult cat-catches. This may be due to dispersal in search of territories or food as their mothers prepare to have new litters.
·We did not see a seasonal relationship between carcass recoveries for avian botulism and depredations. However, we found some interesting information about cats, which may be applied to future management actions such as trapping.
Depredations and Catches
· Depredations showed an increase in the months of April and June, which coincide with the peak waterbird breeding season in Hanalei, but decreased in May. This decrease in depredations coincides with the highest cat-catches and trapping efforts across all months, but it is unclear if the factors are related.
Effort, Catches and Depredations
· Peak cat-trapping effort occurs between November and July when the refuge has the most trained personnel available. As a result, cat-catches are also highest between the months of November and July, showing a positive relationship between trap interval and cat-catches.
· We found that depredation recoveries were not related to cat-catches, and thus depredated carcasses recovered are not a good indicator for cat presence in the refuge.
· There are many confounding factors that influence cat-catches, which points to the need for a cat occurrence (population) index, to monitor efficacy of the cat trapping program.
· We were not able to include the spatial analyses we intended to due to time constraints, however we did figure out the following information:
· Linear length of dike may influence the amount of birds depredated. Taro units account for 75% of the dikes, but disproportionately account for 91% of the depredated carcasses recovered per linear meter in Hanalei NWR. Managed wetlands account for 25% of the dikes, but only 9% of the depredated carcasses recovered per linear meter.
·We did not account for survey effort in these analyses.
·Unpublished USFWS spatial analyses indicates that our highest densities of cats captured on Hanalei NWR occur near suspected cat feeding stations. From 2014 to 2018, we captured 5 notch-eared and microchipped cats on Hanalei NWR in these same areas, indicating the need to better understand feral cat source populations and effects on endangered bird populations.