Breeding range shift of the red-crowned crane (Grus japonensis) under climate change

deviation means the performance of 10 models from 10 fold cross validation is consistent. The predicted current breeding range area is much larger than the actual breeding range of the continental population of the red-crowned crane, suggesting that other factors such as food resource, nesting preference and human built area restrict the breeding range of this species (Fig 1). It is interesting that there are also some suitable breeding areas in the western China and western Mongolia, where is far from the actual breeding area. The overall spatial patterns of the changes of projected breeding area under all four emis- sion scenarios are very consistent under changing climate (Figs 1 and S1, S2 and S3). The breeding range of the continental population of the red-crowned crane will expand northward when project onto future climate change scenarios from our bioclimatic niche modeling. The predicted potential breeding range from our bioclimatic niche model would be contracted as climate change (Figs 1 and S1, S2 and S3). The current actual breeding area is mainly distrib- uted in the center part of Northwest China and along the national border shared by eastern part of Northwest China and Southern Russia. As a result of climate change, the breeding area of continental population would shift to the region along the national border shared by north- ern part of Northwest China and Southern Russia (Figs 1 and S1, S2 and S3). The migratory population of the red-crowned crane, therefore, needs to move northward to track its biocli-

matic niche. There are still some differences among projections of four emission scenarios. The major difference among predictions under the four emission scenarios is that the predic- tions of 2030s, 2050s and 2080s under low emission scenario (RCP2.6) are relatively stable. In addition, the predictions of 2030s, 2050s, 2070s and 2080s under two intermediate emission scenarios (RCP4.5 and RCP 6.0) and high emission scenario (RCP8.5) show that the breeding area would move farther north. We only show RCP2.6, the low emission scenario, in the main text (Fig 1). The latitudinal pattern of projected breeding area of the red-crowned crane by calculating the number of grid cells in 0.5 arc-degree band demonstrated that the breeding area of the con- tinental population of red-crowned crane shift northward under all four RCPs (Fig 2). The current actual breeding area of the red-crowned crane distributes from 43 arc-degree to 54 arc-degree and peaks around 50 arc-degree (Fig 2A). Under the low emission scenario— RCP2.6, the projected breeding areas in 2030s, 2050s, 2070s and 2080s are very similar and peak at 52 arc-degree, suggesting that the breeding area would be stable after 2030s from pre- dictions of our bioclimatic niche model (Fig 2A). The latitudinal patterns of the projected breeding area under RCP4.5 for 2050s, 2070s and 2080s are similar but the size of breeding area would shrink gradually (Fig 2B). The breeding area at 2080s under RCP6.0 is peaking at 55 arc-degree. There are two peaks of breeding area at 2080s under RCP8.5, i.e. 52 degree and

56 degree. The accelerating climate change under higher emission scenario will cause the breeding area of red-crowned crane move farther north. The total number of grid cells predicted to be breeding area of the continental population of the red-crowned crane under RCP2.6 increases at 2030s, and decreases sharply at 2050s, and then increase slightly (Fig 3A). The total number of grid cells under the two intermediate emission scenarios–RCP4.5 and RCP6 maintains the similar pattern of change, it increase at the first, and then decreases afterward (Fig 3B and 3C). The grid number of the predicted breeding area under the high emission scenario—RCP8.5, decreases quickly after 2050s to only half of the number under current climate condition (Fig 3D). The number of predicted grid cells of breeding area of the continental population of the red-crowned crane in both China and Mongolia decreases generally as climate changes over the time periods (Fig 3). For Russia, the number of grid cells predicted to be breeding area for the red-crowned crane under RCP4.5, RCP6.0 and RCP8.5 increases at first and then decreases under changing climate. The number of grid cells in Russia at the end of this century under RCP 2.6, RCP4.5 and RCP6.0 is PLOS ONE Breeding range shift of the red-crowned crane (Grus japonensis) under climate change PLOS ONE | https://doi.org/10.1371/journal.pone.0229984 March 12, 2020 6 / 16