3D scans of Brazilian ceramics from the Calderon Collection (housed at the Museu de Arqueologia e Etnologia in Salvador, Bahia, Brazil) are being used to refine our approach to morphometrics.
Our revised approach will include characteristic ceramic variables using (2D) Elliptic Fourier Analysis (EFA) (using screenshots of the 3D models) and a transition to Landmark Editor for the 3D data (download here), which will allow semi-landmarks to be draped down the vessel profile. We will be abandoning the landmarks used in our pilot study with the exception of CB (center base); the single homologous landmark that will be used across the entirety of the collection, and will drape eight semi-landmark curves down the profile of each vessel to bridge the area between what was RI (rim) and CB – creating four complete profiles of each vessel.
In all, 161 data points were collected from each vessel for the preliminary analysis. Data was exported from Landmark Editor, and imported to Morphologika for generalized Procrustes analysis (GPA) and principal components analysis (PCA). Additionally, PCA1 and Centroid Size were used to demonstrate variation in vessel size within the sample.
We will continue to use Morphologika for GPA and PCA of these 3D datasets, but we are also exploring the geomorph package that is available for R. Preliminary results from the scan data collected in Salvador, Bahia, Brazil (below) point to an area of research with significant potential.
Results of the cluster analysis (again, preliminary) point to four possible shape variants in our sample from Brazil. In this initial run, ceramics from the Aratu tradition (far left – click here to learn more) were found to group together, while vessels from the Tupi-Gurarani tradition (click here to see more) appear to be more variable in shape. Numbers in the graphs below correspond to the numbers in the dendrogram above.
Additionally, we used Morphologika to calculate the regression of PCA1 and Centroid Size to demonstrate variation in size, where the larger vessels appear closer to the right side of the graph and the smaller vessels to the left (below).
We still have plenty of work to do as we seek to refine our method, and are currently working on an elliptic Fourier analysis (EFA) of 2D outlines of these same ceramic vessels to explore how/whether the data from our 3D analysis compares with results of a 2D analysis of harmonics.
Many thanks to our good friends a Geomagic that have collaborated with us on several aspects of this project as we continue to strive for the best possible scan data to use in our analyses.