Data from: Movements of four native Hawaiian birds across a naturally fragmented landscape

Animals often increase their fitness by moving across space in response to temporal variation in habitat quality and resource availability, and as a result of intra and inter-specific interactions. The long-term persistence of populations and even whole species depends on the collective patterns of individual movements, yet animal movements have been poorly studied at the landscape level. We quantified movement behavior within four native species of Hawaiian forest birds in a complex lava-fragmented landscape: Hawai‛i ‘Amakihi (Chlorodrepanis virens), ‘Oma‘o (Myadestes obscurus), ‘Apapane (Himatione sanguinea), and ‘I‘iwi (Drepanis coccinea). We evaluated the relative importance of six potential intrinsic and extrinsic drivers of movement behavior and patch fidelity: 1) forest fragment size, 2) the presence or absence of invasive rats (Rattus sp.), 3) season, 4) species, 5) age, and 6) sex. The study was conducted across a landscape of 34 forest fragments varying in size from 0.07 to 12.37 ha, of which 16 had rats removed using a treatment-control design. We found the largest movements in the nectivorous ‘Apapane and ‘I‘iwi, intermediate levels in the generalist Hawai‛i ‘Amakihi, and shortest average movement for the ‘Oma‘o, a frugivore. We found evidence for larger patch sizes increasing patch fidelity only in the ‘Oma‘o, and an effect of rat-removal increasing patch fidelity of Hawai‛i ‘Amakihi only after two years of rat-removal. Greater movement during the non-breeding season was observed in all species, and season was an important factor in explaining higher patch fidelity in the breeding season for ‘Apapane and ‘I‘iwi. Sex was important in explaining patch fidelity in ‘Oma‘o only, with males showing higher patch fidelity. Our results provide new insights into how these native Hawaiian species will respond to a changing environment, including habitat fragmentation and changing distribution of threats from climate change