Climate change and bird extinctions in the Amazon

deforestation scenarios while D. r. paraensis was profoundly affected in all of the combinations (S1 Fig). Besides, Psophia obscura, which is considered “critically endangered,” according to IUCN [34], may also lose 68.67% and 95.82% of the adequate areas in the mitigation and the business-as-usual scenarios, respectively. Lanio cristatus pallidigula was the only taxon that we did not found future climatically suitable areas in any dispersal contexts or deforestation pro- jections. Thus, this species faces a high risk of being extinct by the year 2050 (Table 2, S1 Fig). We found variations between loss and gain of adequate areas for species in both the opti- mistic and pessimistic predictions. In the mitigation scenario, the average loss of suitable areas for the future is 59%. On the one hand, when analyzing the dispersal contexts in the mitigation framework (GOV), the unlimited dispersal scenario presents an attenuated impact on bird species compared to the other dispersal models, with an average loss of 41.27% of the adequate area. Surprisingly, when we assumed unlimited dispersal, four species have adequate area gains, with Crax fasciolata pinima (+59%) showing the most significant area increment. On the other hand, we found that average area loss was 12% higher when dispersal was limited, even though C. f. pinima, P. paraensis, Dendrocincla merula badia, and Tangara velia signata still showed the same percentages of area expansion in the unlimited dispersal scenario (Table 2, S1 Fig).

The no-dispersal framework shows the most extreme negative results in mitigation projec- tions, where there was an average loss of 82% of suitable areas for the occurrence of the endemic birds, of which eight of them have losses above 80% of their current occurrence geo- graphic range. We also did not found evidence of area expansion for any of the studied taxa (Table 2). Our results indicate that there is no adequate area for D. R. paraensis, Piprites chloris grisescens, and Terenotriccus erythrurus hellmayri in future mitigation scenarios. The business-as-usual scenario shows that even considering three conditions of dispersal (Unlimited, Limited and No dispersal), there was a substantial reduction of adequate area (average: 88%) for all species, i.e., even in a scenario where species can freely disperse, there will be few suitable areas for the establishment of populations. In the business-as-usual Table 2. Estimates of loss and gain of adequate areas in relation to climatic and deforestation impacts, considering the three dispersal frameworks. The mitigation scenario relates the optimistic climate projections (rcp45) to GOV’s predicted deforestation, whereas the Business-as-usual scenario relates the pessimistic climate (rcp85) to the BAU deforestation predictions. The numbers indicate the percentage of variation in the size of suitable areas in the future compared to currently suitable areas. Neg- ative values (—) indicate loss while positive values (+) indicate gains of suitable areas. Taxa Mitigation Scenario Business-as-usual Scenario

Unlimited Dispersal Limited Dispersal No Dispersal Unlimited Dispersal Limited Dispersal No Dispersal Crax fasciolata pinima +59.57% +59.57% -73.54% -99.89% -37.08% -80.24% Psophia obscura -9.45% -97.19% -99.74% -88.22% -99.26% -99.97% Pteroglossus bitorquatus bitorquatus -13.59% -13.59% -73.20% -100% -42.25% -79.74% Piculus paraensis +17.57% -100.00% -100.00% -100% -100.00% -100.00% Thamnophilus aethiops incertus -100.00% -82.07% -85.61% -89.49% -77.26% -81.83% Phlegopsis nigromaculata paraensis -26.46% -28.02% -58.71% -92.96% -69.67% -93.84% Dendrocincla merula badia +14.95% +14.95% -65.66% -99.98% -53.96% -87.56% Dendrexetastes rufigula paraensis -100.00% -100.00% -100.00% -99.72% -100.00% -100.00% Synallaxis rutilans omissa -38.48% -37.97% -59.96% -83.76% -72.08% -79.86% Piprites chloris grisescens -100.00% -100.00% -100.00% -99.10% -100.00% -100.00% Terenotriccus erythrurus hellmayri -100.00% -100.00% -100.00% -98.18% -100.00% -100.00% Ramphocaenus melanurus austerus -28.13% -27.96% -58.00% -83.17% -71.59% -80.46% Granatellus pelzelni paraensis -99.85% -94.07% -99.44% -100% -94.13% -94.13% Tangara velia signata +4.79% +4.79% -67.45% -100% -58.68% -91.73% Lanio cristatus pallidigula -100.00% -100.00% -100.00% -100% -100.00% -100.00% https://doi.org/10.1371/journal.pone.0236103.t002 PLOS ONE Bird extinctions in the Amazon PLOS ONE | https://doi.org/10.1371/journal.pone.0236103 July 17, 2020 8 / 18