INIGEM   23989
INSTITUTO DE INMUNOLOGIA, GENETICA Y METABOLISMO
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Multivariate analysis (PCA) for testing taxonomic assignment in cretaceous conifer woods, Neuquén Basin, Argentina.
Autor/es:
RUIZ, D.; MARTÍNEZ SOLER, M.; PUJANA, R.; MARTÍNEZ, L.C.A.
Lugar:
Mendoza
Reunión:
Congreso; 4° International Palaeontological Congress; 2014
Institución organizadora:
IPA
Resumen:
Fosil woods are traditionally classified by anatomical characters. However, the state of preservation of permineralized woods could not always allow a clear assignment. For this reason, Principal Component Analysis (PCA) would be a useful tool for identification of these fossils. The purpose of this study was to test the use of PCA for corroborating the taxonomic assignment based on anatomical characters; then those characters more easily preserved and useful for systematic classification were selected to be applied in bad preserved fossil woods. The studied specimens are Cretaceous conifer woods from three different localities in the Neuquén Basin, Argentina: Bajada del Overo (Bajada Colorada Formation), Cañadón de las Campanas (Huincul Formation), and Yacimiento Lago Barreales (Portezuelo Formation). Twenty-nine well preserved wood samples were used in this analysis. The woods were previously assigned to two genera (Agathoxylon Hartig and Cupressinoxylon Göeppert) and four species based on anatomical characters. In addition, continuous and discrete characters were selected, based on their relevance as a taxonomic indicator. After performing different PCA, we found that the obtained clusters and the previous taxonomic assignment were compatible. Finally, we performed the PCA to classify the specimens where some characters could not be observed because they were not sufficiently preserved.