Morphological measurements of the Pheidole worker caste in tropical ant communities of Southern Vietnam
Description
The dataset contains morphometric measurements of Pheidole ants collected in Cat Tien National Park (Southern Vietnam) across different forest plots (DFL, L, BS1, BS2, BS3), years (2007, 2008, 2010, 2018), and seasons (wet and dry). For each specimen, head length (HL), head width (HW), Weber’s length (WL), eye length (EL), and hind tibia length (HTL) were recorded, along with caste (minor or major).
Files
Steps to reproduce
Principal Component Analysis (PCA) was employed to compare the morphological traits across different worker subcastes of Pheidole ants. Given that the data did not follow a normal distribution (with the exception of Weber’s length), standardized traits were utilized in the analysis: relative head length (HL/WL), relative head width (HW/WL), relative eye size (EL/HW), and relative length of the hind tibia (HTL/WL). PCA was conducted for the complete species set. The morphofunctional trait space was analyzed using a two-dimensional probability distribution density function, with the X and Y axes representing the first two principal component axes. This resulted in a two-dimensional space that included density contours and a set of density levels, indicating areas where density values exceeded or matched certain thresholds (Chen et al., 2015, 2017). Size Differentiation Analysis: This analysis was performed separately and altogether for different subcastes. The first step involved a correlation analysis using the Kendall’s coefficient, based on the data distribution type. This analysis explored the relationship between Weber’s length and head size (indicators of overall body size) at the subcaste level, with head size calculated as the arithmetic mean of head length and width. Average trait values for each species were used to identify patterns between Weber’s length and head size across subcastes. Kernel Density Estimation (KDE): Niche overlap between co-occurring Pheidole species was quantified using Kernel Density Estimation (KDE). For minors, Weber’s length was used; for majors, head size (mean of HL and HW). Probability density curves were generated with bandwidth selected by the Sheather–Jones method, and overlap was calculated as the intersecting area between distributions. This approach allowed precise quantification of morphological niche overlap and revealed compensatory patterns between worker subcastes.