Figure 4

Relationship between elevation, climates, latitude/longitude, and phenotypic divergence between the lineages. (A) The PCA biplots of elevation and 10 sets of climate data in the 114 Ae. umbellulata accessions shows the factor loadings of each variable (red arrows) and scores of each accession (points) based on the first (PC1), second (PC2), third (PC3), fourth (PC4) components. The analysis employed five monthly climatic data based on WorldClim version 2.1 (Temp: mean temperature, Prec: mean precipitation, Srad: solar radiation, Vapr: water vapor pressure, Wind: wind speed) along with elevation data (Elev), and they were categorized into two seasons: winter season (November to February, labeled "_1") and summer season (March to June, labeled "_2"). Axis labels display the percentage contribution. (B) Scatter plots showing relationship between the PCA scores (PC1 and PC2) obtained from the PCA conducted on the traits (Fig. 2) and day to flowering of the 114 accessions of Ae. umbellulata with respect to latitude/longitude of their habitats. (C) Scatter plots showing relationship between the PC2 scores obtained from the PCA conducted on the traits (Fig. 2) and day to flowering of the 114 accessions of Ae. umbellulata with respect to Solar radiation of winter season (Srad_1). In figures (B) and (C), the day to flowering values in the 2017–2018 season were utilized. A linear regression line, correlation coefficient (r), and p-value of correlation test are shown. In all plots, the colors represent UmbL1e, UmbL1w, UmbL2, and unknown lineages, with blue, yellow, red and gray.