Long-term changes in populations of rainforest birds in the Australia Wet Tropics bioregion: A climate-driven biodiversity emergency

changes. Lowland specialist species have moderately increased in local abundance in the low- lands (+17%). On the other hand, lowland specialists’ local abundance has dramatically increased into the midlands (~190% increase) (Fig 3). Midland specialist species have declined in the lowlands by approximately 42%, by 22% in the midlands and are currently stable in the uplands (Fig 3). Upland specialist species have declined everywhere, with a catastrophic 49% decline in the lower-elevation midlands and a 33% decline in the uplands (Fig 3). Detailed trends for individual species at each elevation category are in the supplementary material (https://doi.org/10.25903/3tmd-sw18). Discussion There is a temptation for policy makers and environmental managers to consider that biodi- versity within a protected area is safe, this is a dangerous, and in this case incorrect, assump- tion. The montane rainforest birds of the Australia Wet Tropics World Heritage Area are in danger of extinction due to the increasingly severe impacts of a changing climate. In 2003, Wil- liams et al. [57] predicted catastrophic extinction levels within the regionally endemic species of the Australian Wet Tropics World Heritage Area by the end of the 21st century. These early predictions were based on bioclimatic envelope models of the changes in species distributions with increasing temperature. However, the reliability of predictions using this simple approach is debateable, with a likelihood of overestimating the severity of impacts [58]. Therefore, a

more intensive monitoring effort was conducted throughout the region, covering ~94% of the environmental space present and providing standardised estimates of local abundance for many species. An earlier analysis using these local abundance measures of rainforest birds across elevation [50] predicted that 74% of rainforest bird species would become threatened by the end of the 21st century. In that analysis, Shoo et al. [50] demonstrated a tight relationship between eleva- tion and the assemblage structure of birds based on empirical, site-based relative abundance across the elevational gradient. Based on this relationship, they predicted that bird assemblages would systematically move upwards in elevation as the climate warmed. However, despite the development of more sophisticated modelling approaches and the availability of more field data, the qualitative outcomes of predictions remained similar to the 2003 predictions. Here, we demonstrate that these predicted trends, whether based on species distribution models or empirical abundance, were concordant with expectations. Our results here demonstrate that Table 1. Multi-species indicator trends between 2000–2016 (overall) and separate trends pre-2009 and post-2008. Overall 2000–2008 2009–2016 Indicator Slope S.E. Trend classification Slope S.E. Trend classification Slope S.E. Trend classification All species 0.998 0.001 Moderate decline 1.003 0.002 Stable 0.982 0.002 Moderate decline Endemic species 0.976 0.001 Moderate decline 0.967 0.003 Moderate decline 0.985 0.004 Moderate decline

Lowland species 1.03 0.002 Moderate increase 1.074 0.005 Strong increase 1.008 0.005 Stable Midland species 0.986 0.001 Moderate decline 0.992 0.003 Moderate decline 0.98 0.003 Moderate decline Upland species 0.971 0.001 Moderate decline 0.952 0.003 Moderate decline 0.972 0.003 Moderate decline Habitat generalists 1.033 0.002 Moderate increase 1.054 0.005 Moderate increase 0.974 0.005 Moderate decline Rainforest specialists 0.983 0.001 Moderate decline 0.981 0.002 Moderate decline 0.986 0.002 Moderate decline Population trends were classified following Pannekoek and van Strien [56] into one of the following categories depending on the overall multiplicative slope and its 95% confidence interval. Strong increase/Steep decline: increase/decline significantly more than 5% per year. Moderate increase/decline: significant increase/decline, but no more than 5% per year. Stable: no significant increase or decline, and it is certain than trends are less than 5% per year. Uncertain: no significant increase or decline, but it is not certain if trends are less than 5% per year. https://doi.org/10.1371/journal.pone.0254307.t001 PLOS ONE Rainforest birds on escalator to extinction PLOS ONE | https://doi.org/10.1371/journal.pone.0254307 December 22, 2021 5 / 16