Classification of some waterlogged variety testing environments on Ethiopian vertisols on the basis of grain yield response of Faba Bean genotypes.

Information on varietal discrimination patterns across the testing environments facilitates the design of effective spatial and temporal replication of variety trials and a more efficient use of resources. Data from variety trials of 12 faba bean genotypes grown between 1997 and 2000 at multi-locations on adjacent fields of broadbeds and furrows (BBF) and flat beds on waterlogged Vertisols were used to classify the testing environments. Differences were significant among the environments and genotypes (G x E) and due to interaction effect. Clustering grouped the genotypes into two broad categories indicating that they contrast for yield potential. The environments were grouped into three clusters each for BBF and all possible environments and two each for flat bed and for mean performance across BBF and flat bed. The locations in the northern (Enewari and Bichena) part of the country and those in the central and south-eastern (Ginchi, Ambo and Sinja) parts showed a tendency to be grouped together regardless of years and moisture management technique. Enewari 1998, Bichena 1999 and Ginchi 1999 and 2000 all on BBF did not follow this general trend indicating that geographic position was not the only factor influencing clustering. The classification of locations into two broad categories suggests t he need for a separate primary breeding site or testing location or each of the categories commencing from an early stage in the breeding programme. Therefore, inter-cluster spatial replication of trails is more important than replication targeted to capture intra-cluster variation, temporal differences and those due to moisture management technique. However, a comprehensive study involving both edaphic and climatic factors with a greater number of locations representative of the Vetisols of the Ethiopian highlands is needed for a more tangible conclusion at the national level.