Fig 1.
Distribution of Welwitschia mirabilis in Namibia and location of the study area.
In the main map, the black polygon indicates the study area, the red polygons show the known species distribution, and the blue polygons represent the boundaries of the observed extent of occurrence in Northern Namibia. In the inset map the red circle indicates the general location of the study area in Africa. Map tiles by Stamen Design, under CC BY 3.0.
Fig 2.
Historical and future climatic conditions in the welwitschia extent of occurrence.
Density plot of historical data (in green) and expected future values (in orange) for annual mean temperature (A) and annual precipitation (B) in the welwitschia extent of occurrence. Density plots use kernel density estimates to show the probability density function of the variables.
Fig 3.
Observed variability of plant traits.
Density plots of measured values for: proportion of plants in poor conditions (A), proportion of plants in average conditions (B), proportion of plants in good conditions (C), proportion of plants with cones (D), proportion of dead plants (E), sex ratio (F), stem major axis (G), stem minor axis (H), and leaf length (I). Density plots use kernel density estimates to show the probability density function of the variables.
Fig 4.
Variation of climatic suitability for W. mirabilis in the study area.
Expected suitability variation from climate change (red and green shades indicate negative and positive variations respectively) calculated as the difference between future and historical suitability. Black polygon indicates the study area, and the blue polygon represents the boundaries of the observed extent of occurrence in Northern Namibia.
Fig 5.
Linkage between population features and the expected suitability variation.
Plots of the proportion of plants in poor (A), average (B), and good conditions (C), the proportion of plants with cones (D), the proportion of dead plants (E), the number of plants (F), the stem major (G) and minor axis (H), and the leaf length (I) as functions of the expected suitability variation in the stands. In all the plots, blue dots are values for plant stands, colored lines and areas represent the regression lines and the confidence intervals respectively, and the texts on the top of the plots indicate the results of the null-model correlation analyses (Pearson r and corrected p values for multiple comparisons). Note that suitability variation values (X-axis) are all negative; thus, the reduction of suitability increases from right to left.
Table 1.
Linkage between population features and the stand latitude and altitude.
Results of the null-model correlation analyses (Pearson r and corrected p values for multiple comparisons).