New Zealand Journal of Ecology (2018) 42(2): 149- 157

Changes in density of hihi (Notiomystis cincta), tīeke (Philesturnus rufusater) and tūī (Prosthemadera novaeseelandiae) on Little Barrier Island (Te Hauturu-o-Toi), Hauraki Gulf, Auckland, 2005–2013

Research Article
Robin Toy 1*
Terry C. Greene 2
Brenda S. Greene 3
Alicia Warren 4
Richard Griffiths 5
  1. 78c Little Sydney Road, Motueka 7198, New Zealand (E-mail:
  2. Department of Conservation, PO Box 13049, Christchurch 8141, New Zealand
  3. Christchurch City Council, PO Box 73011, Christchurch 8154, New Zealand
  4. Department of Conservation, Private Bag 68908, Newton, Auckland 1145, New Zealand
  5. 561 Woodcocks Rd, RD1, Warkworth 0981, New Zealand
*  Corresponding author

The hihi/stitchbird (Notiomystis cincta), an endangered New Zealand endemic species, has one self-sustaining population not subject to human intervention, located on Little Barrier Island (Te Hauturu-o-Toi), in the Hauraki Gulf, Auckland. All other hihi populations have been derived from Little Barrier Island and all require active management. Changes in the population of hihi on Little Barrier Island are, therefore, of great conservation interest. During 2005–2013, densities of hihi were compared using distance sampling to those of two other endemic forest birds: tūī (Prosthemadera novaeseelandiae) and tīeke/North Island saddleback (Philesturnus rufusater). Tūī has a broadly similar diet with both adapted to nectivory, and tīeke a more omnivorous diet. During 2005–2006, hihi densities were relatively high (3.1–4.0 hihi ha-1) but declined during 2007 (1.3 hihi ha-1). Similarly, tūī had comparatively high densities during 2005–2006 (3.0–4.4 tūī ha-1 respectively) but declined during 2007 (1.6 tūī ha-1). By 2009, hihi density increased (3.1 hihi ha-1) then declined and remained at low densities during 2010–2013 (0.8–1.1 hihi ha-1). Tūī density also increased by 2009 (2.2 tūī ha-1), but was variable during 2010–2013 (0.7–3.3 tūī ha-1). In contrast, tīeke densities remained relatively stable from 2005–2013 (1.7–2.8 tīeke ha-1). Extrapolating from 2013 estimates of 1.0 hihi ha-1, 1.5 tūī ha-1 and 2.2 tīeke ha-1, the island populations were about 3100 hihi (95% confidence interval (CI) 2500–3400), 4600 tūī (95% CI 4300–4900) and 6800 tīeke (95% CI 6200–7400). The variability in hihi and tūī densities suggests strong drivers of population change, perhaps availability of preferred food types. However, the comparatively constant hihi density in the latter half of the study compared to more variable tūī density suggests other factors are affecting the tūī population. The lower variability in tīeke density suggests their populations are influenced by other factors. Notwithstanding historical natural variation in hihi density, the low densities recorded during 2010–2013 are likely to have increased the risk of loss due to both short-term stochastic and long-term environmental change. We therefore recommend continued monitoring of the hihi population using distance sampling and investigation of the factors that influence hihi density.