Flesh-footed shearwater decline on Lord Howe: Rebuttal to Lavers et al. 2019

Abstract

https://doi.org/10.1016/j.gecco.2019.e00794 2351-9894/© 2019 The Authors. Published by Elsev licenses/by-nc-nd/4.0/). We refer to ‘Changes in technology and imperfect detection of nest contents impedes reliable estimates of population trends in burrowing seabirds’ by Lavers et al. (2019). This paper reports on a recent (2018) survey of the threatened Flesh-footed Shearwater (Ardenna carneipes) on Lord Howe Island. The 2018 estimate is then compared with estimates from three previous surveys (in 1978, 2002 and 2009) to examine population trends. The paper s resulting conclusion, “that the Flesh-footed Shearwater population on Lord Howe has decreased by up to 50% in the last decade” we believe to be both fallacious and unjustified based on the data presented. The 2018 survey used a similar approach to that of the previous three surveys (Fullagar and Disney, 1981; Priddel et al., 2006; Reid et al., 2013): mapping burrow distribution, counting burrows to determine burrow density, and inspecting burrows using a burrowscope to determine occupancy rate. The exception being the 1978 survey that did not assess occupancy rate, but assumed an occupancy rate of 50% (an approximation based on data from seabird colonies worldwide). Lavers et al. (2019) estimate the population of Flesh-footed Shearwater on Lord Howe Island to be 22,654 (range 8156e37,909) breeding pairs in 2018, an estimate that, given the substantial errors involved, is roughly consistent with those from the three previous surveys (Table 1). If anything, the 2018 data indicate a recent population increase, along with an increase in colony area, burrow number and burrow density. Contrary to the data, Lavers et al. (2019) conclude that the population has declined by up to 50% in the last decade. They base this conclusion on their belief that “previous surveys likely underestimated true occupancy because of technological challenges”. They justify this belief by comparing the effectiveness of two different versions of burrowscope; one of which is obviously fit-for-purpose (i.e., it provides a clear image of the contents of the burrow), the other is obviously not (i.e., the image is inadequate to determine the contents of the burrow; see Figure 2 in Lavers et al., 2019). Without explanation or justification, Lavers et al. (2019) presume that the two previous surveys (2002 and 2009) utilised burrowscopes comparable to the one that they demonstrated to be not fit-for-purpose (Sextant Technologies scope). This assumption is incorrect. Both previous surveys used bespoke burrowscopes built by skilled technicians at university workshops, using the best available components sourced from around the globe. These burrowscopes may have relied on monochrome monitors of lower resolution thanwhat is available today, but the images produced were no less informative in assessing burrow occupancy. The only downside to these devices was the excessive bulk and weight of the power source. We concede that occupancy rates could not be determined with complete accuracy, but this, as always, was due to the structure