Populations and ecosystems can have multiple stable states, separated by a tipping point.   Systems far from the tipping point recover quickly (high resilience); near the tipping point recovery is slower (low resilience).  Slow recovery associated with a tipping point, or critical slowing down, can be preceded by early warning signals, including increases in autocorrelation, variance, and skewness as the system moves towards transition.

While the conservation applications of recognizing these signal are commonly discussed, early warning signal detection methods are rarely applied to ecological data and their validity and utility in conservation management remain unclear.

There is debate about the current population trends and predicted short term fates of the endangered forest birds Hawaii Creeper (Loxops mana)  and Hawaii ‘Akepa (L. coccineus) in Hakalau Forest NWR, Hawaii. Some studies report the populations as stable or increasing (1-3) while others  report signs of decline and imminent population collapse associated with the rapid increase of an invasive competitor, the Japanese White-eye (Zosterops japonicus), starting in 2000 (4,5).

If there is a recent or impending transition in creeper and ‘akepa populations, from persistent dynamics to dynamics associated with imminent extinction, it might be statistically detectable by early warning signals of critical transition.

I presented this research at the 2016 North American Ornithological Conference in Washington, DC in August.



1. Camp et al. 2010 Condor 112: 196-212
2. Camp et al. 2014 Condor 116: 97-101
3. Camp et al. 2016 Bird Conservation International 26: 225-242
4. Freed & Cann 2010 Condor 112: 213-221
5. Freed & Cann 2013 Condor 115: 442-447