Predicting the distributions of Pouteria adolfi-friederici and Prunus africana iconic tree species under current and future climate change scenarios in Ethiopia

The distributions of the potential adaptive ranges of iconic plant species are not yet fully known especially in regions such as Ethiopia where high climatic variability and vegetation types are found. This study was undertaken to predict the distributions of the potential suitable habitats of Pouteria adolfi-friederici and Prunus africana tree species under the current and two future climate scenarios (i.e., RCP 4.5 and RCP 8.5 in 2050 and 2070) using MaxEnt software (version: 3.4.4.). Eleven less correlated environmental variables (r<0.7) were identified and used to make the prediction models. Elevational shifts of the highly suitable habitats, effects of elevation, solar radiation and topographic position in relation to the current and future climatic scenarios on the habitats were statistically analyzed using independent t-test and linear model. Under all climate scenarios, we found a decrease in the proportion of areas of highly suitable habitats for both study species. High potentials of suitable habitats for Pouteria adolfi-friederici are predicted to be confined to southwest, west central and south parts of Ethiopia in fragmented moist afromontane forest patches, while it is in the southwest and west central parts of Ethiopia for Prunus africana. On basis of vegetation types of the country, potential suitable habitats for Pouteria adolfi-friederici are predicted to occur in moist evergreen forest, dry evergreen forest and grassland complex and Combretum-Terminalia woodland vegetation types. Whereas, moist evergreen forest, dry evergreen forest and grassland complex and riverine vegetation types are predicted to comprise potential suitable habitats for Prunus africana, showing considerable spatial dynamics. Overall, our results suggest that the strategies deem to design biodiversity conservation should take into account the dynamics of the suitable niches of different species under different future climate scenarios.


52
Plant species have their optimum range of climatic condition to survive in their ecological zones 53 [1]. Among others, the environmental factors limiting the adaptive range of the plant species are 54 altitude, temperature, moisture, and soil types [2,3]. Climate, which represents a weather  Table 1) for conservation 115 purposes, and we have used them for this study. The species occurrences data of both species were collected from the National Herbarium of 119 Ethiopia and from GBIF (www.gbif.org). Moreover, geographical coordinates of species presence 120 locations were collected using a handheld Garmin GPS in west central areas of the country 121 during a field work. In total, 94 coordinates (i.e., 30 for Pouteria adolfi-friederici and 64 for 122 Prunus africana) points were used in this study. 123 As the current input data, the 19 bioclimatic variables (1970−2000) with a spatial resolution of 124 about 1 km 2 were downloaded from www.worldclim.org/data/worldclim21.html. Furthermore, 125 solar radiation index (sri) was downloaded from www.worldclim.org/data/worldclim21.html  (Table in S1 Table). Eleven variables that are less correlated (r<0.7) were included in prediction 140 models and this cut-off point was commonly applied to fit models to bioclimatic data [50−52].

141
The 11 bioclimatic variables used in the prediction models are mean monthly temperature, 142 isothermality, temperature seasonality, mean temperature of coldest quarter, precipitation   Quarter and 500−700 mm Precipitation of Warmest Quarter (Fig 2).    suitable habitats cover most of the central, eastern and western highlands of Ethiopia ( Fig 5). Under the current climatic conditions, the less suitable and moderately suitable habitats cover 220 larger areas compared to highly suitable and suitable habitats for both Pouteria adolfi-friederici 221 and Prunus africana (Fig 6 and 7). suitable habitats (Fig 4 and 5). Unlike these three suitability classes, the relative area of the 235 highly suitable habitats for both tree species does not significantly vary across these scenarios 236 (Fig 6 and 7).   The highly suitable habitats of Prunus africana tend to decline from the current to under future 252 climate change scenarios in 2050 and 2070 (Fig 9). These habitats class will be useful for this 253 species in CTW until 2050 for RCP 4.5, but it will be fully non-existent in 2070 and for RCP 8.5.

254
There will also be suitable habitats in DAF for Prunus african despite DAF will be with less      Interestingly, the prediction models show that the distributions of the suitable habitats or niches 295 of these tree species can shift among the vegetation types or ecosystems due to the effect of the 296 future climate change scenario (Fig 4 and 5 Our study has revealed that the environmental factors such as, solar radiation, Precipitation of 304 Warmest Quarter and elevation have significantly mediated the distributions of the potential 305 suitability habitats of the two study species (Table 2)  The highly suitable areas of Prunus africana in Combretum-Terminalia was identified for RCP Terminalia and dry afomontane vegetation types (Fig 5). However, these habitats seem 345 unfavorable for this species under RCP 8.5 may be due to the impact of high temperature.

346
The spatial distributions of potential suitable habitats of Pouteria adolfi-friederici and Prunus 347 africana tree species showed significant shifts along elevational gradients under the current and 348 the future scenarios (Table 3). Similarly, topographic position was found to be important factor 349 in driving the distributions of the suitable habitats of Pouteria adolfi-friederici and Prunus 350 africana tree species, but such impact was not found for elevation and solar radiation (Table 4   351 and 5).

352
In general, in terms of the relative proportions of the areas predicted to be suitable habitats for       Table 1) for conservation 709 purposes, and we have used them for this study.  Table). Eleven variables that are less correlated (r<0.7) were included in prediction models and  (Table 2).  quarter and 500−700 mm precipitation of warmest quarter (Fig 2).  (Fig 3).

811
Under the current climatic conditions, the less suitable and moderately suitable habitats cover 812 larger areas compared to highly suitable and suitable habitats for both Pouteria adolfi-friederici 813 and Prunus africana (Fig 6 and 7). suitable habitats (Fig 4 and 5). Unlike these three suitability classes, the relative area of the 825 highly suitable habitats for both tree species does not significantly vary across these scenarios 826 (Fig 6 and 7). The highly suitable habitats of Prunus africana tend to decline from the current to under future 841 climate change scenarios in 2050 and 2070 (Fig 9). These habitats class will be useful for this 842 species in CTW until 2050 for RCP 4.5, but it will be fully non-existent in 2070 and for RCP 8.5.

843
There will also be suitable habitats in DAF for Prunus african despite DAF will be with less     Interestingly, the prediction models show that the distributions of the suitable habitats or niches 883 of these tree species can shift among the vegetation types or ecosystems due to the effect of the 884 future climate change scenario (Fig 4 and 5 (Fig 8 and 9) woodland vegetation type (Fig 4 and 8).

931
The highly suitable areas of Prunus africana in Combretum-Terminalia was identified for RCP Terminalia and dry afomontane vegetation types (Fig 5 and 9). However, these habitats seem to 934 be unfavorable for the species under RCP 8.5, may be due to the impact of high temperature that 935 is assumed to increase under this trajectory than the current and RCP 4.5 scenarios.

936
The spatial distributions of potential suitable habitats of Pouteria adolfi-friederici and Prunus 937 africana showed significant shifts along elevational gradients under the current and future 938 climate scenarios (Table 3). Similarly, topographic position was found to be important factor in 939 driving the distributions of the suitable habitats of Pouteria adolfi-friederici and Prunus africana 940 tree species, but such impact was not found for elevation and solar radiation (Table 4 and 5).

941
In terms of the relative proportions of the areas predicted to be suitable habitats for both tree 942 species, the trend show that the areas are decreasing under future climate scenarios when 943 compared with the current status (Fig 6 and 7). In this connection, our results corroborate the 944 earlier findings that pointed out that climate change is highly responsible for the area contraction