FIGURE LEGENDS
Figure 1 Occurrence locations of Hypsignathus monstrosus sightings in red (123 observations in Africa of which 9 in DR Congo), Myonycteris torquata sightings in blue (79 observations in Africa of which 7 in DR Congo) and Epomops franqueti sightings in green (201 observations in Africa of which 10 in DR Congo)
Figure 2 Receiver Operating Characteristic (ROC) curve and Area Under the Curve (AUC) value under the current period (100 runs) forEpomops franqueti (A), Hypsignathus monstrosus (B) andMyonycteris torquata (C).
Figure 3 Jackknife test for variable importance of Hypsignathus monstrosus (A) , Myonycteris torquata (C) and Epomops franqueti (B) . Mean Diurnal Range (bio2), Min Temperature of Coldest Month (bio6), Annual Temperature Range (bio7), Temperature of Wettest Quarter (bio8), Mean Temperature of Driest Quarter (bio9), Mean Mean Temperature of Warmest Quarter bio10), Precipitation Seasonality (bio15), Precipitation of Wettest Quarter (bio16), Precipitation of Driest Quarter (bio17), Precipitation of Warmest Quarter (bio18), Precipitation of Coldest Quarter (bio19).
Figure 4 Response curves of the variables contributing most to the prediction by the MaxEnt models for Epomops franqueti, Hypsignathus monstrosus and Myonycteris torquata. Red indicates the mean response averaged over the 100 replicate MaxEnt runs, while blue indicates one standard deviation. For Epomops franqueti, bio7 (Annual Temperature Range) in Figure 4A had the largest percent contribution and the largest permutation importance, followed by bio17 (Precipitation of Driest Quarter) in Figure 4B. For Hypsignathus monstrosus, bio7 (Figure 4C) had the largest percent contribution; bio17 (Figure 4D) and bio19 (Precipitation of Coldest Quarter) in Figure 4E had the largest permutation importance. For Myonycteris torquata, bio17 (Figure 4F) and bio19 (Figure 4G) had the largest permutation importance.
Figure 5 Predicted current distribution of bats in DR Congo. The letters A, B, C, D in the maps represent the predicted distribution ofEpomops franqueti, Hypsignathus monstrosus, Myonycteris torquataand mean relative probability of presence respectively. Dark green indicates not suitable area (0.00 - 0.05); Light green indicates slightly suitable area (0.05 - 0.25); Yellow indicates moderately suitable area (0.25 - 0.50); Orange indicates suitable area (0.50 - 0.75); Red indicates highly suitable area (0.75 - 1.00).
Figure 6 Predicted distribution of bats in DR Congo in 2050 under RCP 4.5 (top) and RCP 8.5 (bottom). The letters A, B, C, D in the maps represent the predicted distribution of Epomops franqueti, Hypsignathus monstrosus, Myonycteris torquata and mean relative probability of presence respectively. Dark green indicates not suitable area (0.00 - 0.05); Light green indicates slightly suitable area (0.05 - 0.25); Yellow indicates moderately suitable area (0.25 - 0.50); Orange indicates suitable area (0.50 - 0.75); Red indicates highly suitable area (0.75 - 1.00).
Figure 7 Predicted distribution of bats in DR Congo in 2070 under RCP 4.5 (top) and RCP 8.5 (bottom). The letters A, B, C, D in the maps represent the predicted distribution of Epomops franqueti ,Hypsignathus monstrosus , Myonycteris torquata and mean relative probability of presence respectively. Dark green indicates not suitable area (0.00 - 0.05); Light green indicates slightly suitable area (0.05 - 0.25); Yellow indicates moderately suitable area (0.25 - 0.50); Orange indicates suitable area (0.50 - 0.75); Red indicates highly suitable area (0.75 - 1.00).