The species inhabits remnant native forests, on which it appears to be completely dependent for breeding (there is a strong negative correlation between the presence of non-native vegetation and the location of nests [P. Hodum in litt. 2007, 2008]). However, throughout the year it also utilises non-native plant communities, feeding on introduced plants, such as Eucalyptus globulus and garden flowers. This usage of non-native plants is especially common in the austral autumn and winter when only one native species (Raphithamnus venustus) flowers (Roy et al. 1999). It feeds on the endemic Cabbage Tree (Dendroseris litoralis) in late summer (Hahn et al. 2015). It is mostly nectarivorous, but small insects are taken from leaves or in flight. The proportion of insects in the diet increases during the chick-rearing period. The sex ratio is heavily skewed, with three males to every female (Roy et al. 1999). It may experience competition with S. sephaniodes, especially over access to Dendroseris litoralis flowers post-breeding, which may lead to spatial segregation during the post-breeding season (López-Calleja et al. 2006, Vizentin-Bugoni et al. 2017).

Sephanoides fernandensis is endemic to the Juan Fernández Islands, Chile, where the nominate race is confined to c. 11 km² on Isla Robinson Crusoe (Hahn et al. 2005). The race leyboldi is extinct on Isla Alejandro Selkirk, where it was last recorded in 1908.

Conservation Actions Underway
CITES Appendix II. The Juan Fernández Islands were designated as a national park in 1935 (protected from 1967) and an UNESCO Biosphere Reserve in 1977. The Chilean government began restoring habitat in 1997 (J. C. Torres-Mura in litt. 1999), and the islands have been nominated for World Heritage listing (Hulm and Thorsell 1995). Conservation is being led by Oikonos (Oikonos, n. d.). Key activities which have already taken place are (Hodum and Rodriguez 2005, P. Hodum in litt. 2007, 2008): the hiring of two island residents as project coordinators; control of invasive plants and herbivores (including volunteer programmes for island residents to participate in invasive plant removal), which appears to increase nesting success of the species; invasive predator control (including cat control in the town on Robinson Crusoe); habitat restoration in native forest; a community outreach programme aimed at engaging local people (Anon. 2005) and including environmental education programmes for local schoolchildren; as well as population surveys and monitoring of active nests, phenology and reproductive success. The endemic plant Dendroseris litoralis has been planted in settlement areas to provide nectar resources for the species (Hahn et al. 2011b). The population has been monitored since 2006, and a workshop on terrestrial bird census techniques was organized for local people in October 2011, aimed at forming a local team that can carry out regular bird censuses; a pilot project was implemented to test the capacity of the trained team of local fieldworkers to gather data by using a monitoring protocol. Ongoing and planned work includes further invasive plant control, the planting of native plant species, carrying out environmental education workshops in the local high school and the development of a programme of cat control (Oikonos n.d., Mohamed bin Zayed Species Conservation Fund 2017). A national recovery strategy is being prepared by the Chilean Ministry of the Environment, assisted by Oikonos (P. Hodum in litt. 2020).

Conservation Actions Proposed
Continue to monitor the population using a quantitative census methodology that allows for statistical comparisons between surveys. Carry out mark-recapture studies to describe key demographic parameters needed for conservation management action (E. Hagen in litt. 2020). Investiage whether the competition between S. fernandensis and S. sephanoides during the non-breeding season impact densities of S. fernandensis (J. Vizentin-Bugoni in litt. 2020). Take measures to control S. sephanoides if negative impacts on S. fernandensis exist (J. Vizentin-Bugoni in litt. 2020). Survey additional areas of high-quality forest to determine if other areas of critical breeding habitat exist (P. Hodum in litt. 2020). Ensure that Cabbage Trees that were destroyed by the 2010 tsunami are replanted (Hahn et al. 2015). Enforce grazing restrictions on National Park land (Roy et al. 1999). Evaluate feasibility of establishing feeding stations in native forest. Continue to replant native plants that are critical food sources (including Sophora fernandeziana and Rhaphithamnus venustus), especially in watersheds far from San Juan Bautista where the highest densities of cats are concentrated (E. Hagen in litt. 2020, P. Hodum in litt. 2020). Continue to support ongoing efforts and expand them further to remove alien invasive plants and mammalian predators, and increase awareness. Investigate scale of predation risk from Austral Thrush and consider control measures if appropriate (Hahn et al. 2011a). Evaluate genetic variation and inbreeding depression within the population (Roy et al. 2013).