Degree Type


Date of Award


Degree Name

Doctor of Philosophy




Iowa counties were grouped into agroclimatic regions by numerical taxonomic techniques (cluster analysis). The algorithm used was Ward's method (hierarchical method). The method used squared Euclidean distance as a similarity coefficient;Standardized variables were obtained for each of 30 environmental variables (mean monthly temperature, maximum monthly temperature, minimum monthly temperature, and monthy precipitation each from May to October; annual precipitation, growing degree days, frost free season, first Fall frost, average stress index, and plant available soil moisture capacity). The data set used in this study included means of 30 years (1951-1980) for each of 95 Iowa counties for the 30 variables;The cluster analysis (Ward's method) classified groups generally in close geographic proximity, even though geographical variables (latitude, longitude, etc.) were not used in the clustering process;The results of the analysis indicated that the method of cluster analysis (Ward's method) was acceptable when grouping the counties of Iowa into five regional agroclimatic zones by 30 climatic variables. Those regions were: Northwest, Northeast, Central, South, and a transitional region between central and south regions;The counties were grouped according to statistically derived relationships. Five major groupings were identified for Iowa. These groupings are expected to provide a more realistic basis for selecting experimental sites to avoid large cultivar x environment interaction when testing corn hybrids;An agroclimatic classification was produced for the state of Iowa that is complementary to earlier efforts in grouping the soil of the state. When considered together, the soil associations and climate regional groupings correlated well with observed corn yield obtained over a 30-year period of record;Methods were developed to enable prediction models of crop suitability to be generated for locations (states or nations) where intensive cropping practices are not well-developed.



Digital Repository @ Iowa State University,

Copyright Owner

Marcial Ortiz-Valdez



Proquest ID


File Format


File Size

104 pages