Classification of maize testing locations in Sub-Saharan Africa by using agroclimatic data

Classification of locations in CIMMYT's International Maize Testing Program in Sub-Saharan Africa into regions was done by using agroclimatic variables. The study was undertaken to improve the efficiency of the testing program. Agroclimatic variables used to classify locations included altitude, average monthly and total yearly rainfall, average monthly temperatures, average monthly minimum temperatures, average monthly maximum temperatures, and average yearly minimum and maximum temperatures. Averages were calculated over a minimum of 10 years' data. The classification procedures used were Ward's method of cluster analysis and canonical discriminant analysis. Good discrimination of regions of maize testing locations was achieved by using all the variables, and cluster analysis resulted in groups that can be defined as two areas of West Africa with different rainfall patterns, and three areas of East and South Africa withdifferent temperature patterns. Final determination of the efficiency and use of these regions and whether all locations were correctly classified will be determined by using input of collaborators with personal experience at the locations and actual practice in the International Maize Testing Program.

Bibliographic Details

Main Authors:	Pollak, L.M., Pham, H.N.
Format:	Article biblioteca
Language:	English
Published:	Consiglio per la Ricerca e la sperimentazione in Agricoltura, Unità di Ricerca per la Maiscoltura 1989
Subjects:	AGRICULTURAL SCIENCES AND BIOTECHNOLOGY, Cluster Analysis, Canonical Discriminant Analysis, ZEA MAYS, VARIETY TRIALS, STATISTICAL METHODS,
Online Access:	http://hdl.handle.net/10883/3338
Tags:	Add Tag No Tags, Be the first to tag this record!