Application of Multivariate Analysis on Morphometric Traits to Characterize the Sheep Populations Reared in the Central Rift Valley of Ethiopia

Abstract

The study was conducted in Adam-Tullu Jiddo-Kombolcha (ATJK), Bora, Asasa and Kofele districts situated at the Central Rift Valley of Oromia, to describe the morphological diversity of sheep populations by applying multivariate analysis. Morphometric traits were taken from 523 ewes of various age groups. The traits scored were live weight (LW), body length (BL), height at withers (WH), heart girth (HG), height at rump (RH), chest depth (CD), chest width (CW), rump length (RL), rump width (RW), head length (HL), head width (HW), ear length (EL), canon circumference (CC), canon length (CL), tail length (TL), tail circumference (TC) and neck length (NL). Results revealed that district had a highly significant effect on most traits except RL, RW, and HL. The age effect was also significant for all traits except CC, CL and TC. The sheep of Kofele had the highest LW, BL and TL values while those of the Asasa showed the lowest RH, and WH being the shortest amongst the studied sheep populations. The highest WH was noted in Kofele sheep being the tallest (p<0.01) from Asasa and Bora sheep. The longest EL was observed in Asasa and Kofele sheep being higher (p<0.01) than those of ATJK and Bora. Most morphometric traits increased with advancing age. Body length, TC, CW, TL and LW were identified as the most discriminating variables to differentiate the four district sheep populations. All Mahalanobis distances were significant (p<0.01) being the shortest between sheep of ATJK and Bora (0.76) while the longest was observed between those of Asasa and Bora (3.60). Three canonical variables (CAN) were determined of which CAN1 and CAN2 accounted for 69.0 and 27.0% of the total variations, respectively. The CAN1 loaded highly for RH while CAN2 weighted for CD. About 72.3%, 68.6%, and 64.7% of original sheep were correctly classified into their respective source population of Asasa, Kofele and Bora districts. However, only 36.4% of the original ATJK sheep were assigned into their respective population while the remaining being misclassified to other sheep populations. The present work revealed that characterization of indigenous livestock based on morphological traits using a multivariate analysis is a viable option in regions where molecular tools are inaccessible. We recommend initiation of a community-based breeding program for sustainable utilization and conservation of Kofele sheep.