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

Lowland species are mostly increasing in abundance thus far with increases in abundance in the lowlands and strong increases in range and abundance into the midlands (Fig 3 and Table 1). The pattern in temporal change in abundance of rainforest birds presented in this study is consistent with observed species elevational shift worldwide [22,33,40,41,59–61]. Globally, evi- dence suggests a strong uphill elevational shift across different taxa, indicating that this trend could be a generality. However, some studies have pointed out some idiosyncratic results found in species-specific trends within different communities [33,59]. According to this, our results showed that, even though the uphill distribution shift is marked across most species, some birds showed stable trends, while others showed pattern opposite than expected (e.g., Acanthiza katherina). Those individual exceptions to the overall pattern highlighted potential interspecific differences in dispersal capacity, resilience, and adaptability between bird species in the Wet Tropics, whose impact in the overall community “re-shuffling” at the ecosystem level should be further studied. Overall, this study supports growing evidence of the rapid impact that climate change is having on tropical ecosystems [34] and provides the first evi- dence of a climate-induced elevational shift in tropical rainforest of Australia. So, what is driving these shifting abundance patterns? Is it the direct impacts of temperature on the birds or is it an indirect effect via food resources or other biotic interactions? There is

little confidence in our ability to predict the potential impacts of the complex changes in biotic interactions due to these marked changes in abundance and geographic shifts in distributions. Upland and midland assemblages are likely coming under increased pressure due to changing biotic interactions and community structure associated with the increasing abundance of lower elevation species as they shift higher up the mountain. The changes shown in this study suggest that, thus far, the impacts on species have been largely direct, or indirect via changes in availability of an important resource, rather than changes in competitive interactions between bird species. We argue this on the basis that upland bird species have declined in their core range (>850m) despite there being no significant increase in midland birds in the uplands thus far. Midland birds have only declined by about 20% in the midlands despite the influx of lowland species. Lowland species have increased (albeit with more recent trends of decline) and there has been no noticeable change in species composition or new invading species in the lowlands to date. Thus far, there is no evidence supporting “lowland attrition” of species in this system [37]; however, this is not entirely surprising given that we have little knowledge of the upper thermal limits of the lowland species because they already occur in the hottest part of the region Shoo, Anderson [62] suggested that the elevational pattern of abundance for the Lewin’s Honeyeater was directly influenced by temperature and not competitive interactions.

This conclusion was based on a demonstration that the elevational abundance profile of the species varied as predicted by temperature in the populations on mountains on Cape York, a hotter montane system, within a very different bird assemblage to that in the Wet Tropics region [62]. However, this does not account for other biotic interactions that might also be changing such as diseases, or other non-bird competitors such as ectotherms (invertebrates mostly) that might be increasing in uplands. There are many indirect mechanisms that could potentially exacerbate the impact that cli- mate change will ultimately have on bird species. Increased pressure from parasites with increasing temperature [63], decreased reproductive success during prolonged dry seasons [64,65], increased habitat and population fragmentation [66–68] and increased environmental marginality [69]. In this region, Williams, Shoo [55] hypothesised that dry season bottlenecks and changes in net primary productivity could have a strong influence on bird populations. Net primary productivity is limited by both temperature and water availability and could pos- sibly increase in the uplands with a warming climate, potentially alleviating some of the more direct impacts of high temperatures. However, the strong declines in the abundance of upland PLOS ONE Rainforest birds on escalator to extinction PLOS ONE | https://doi.org/10.1371/journal.pone.0254307 December 22, 2021 8 / 16