A synoptic approach for crop loss assessment used to study wheat. I. An appraisal of the physical and chemical soil properties in the study area

Abstract

A synoptic approach to crop loss assessment is proposed whereby the relative importance of various factors in the environment can be assessed in relation to the growth and yield of plants. A degree of simplification of complex sets of data, collected in field surveys, can be achieved by analysing natural subsets of the data using multivariate statistical procedures. The concepts that emerge form a basis for constructing regression models to predict growth and yield by influencing the selection of variables for inclusion that ensure statistical stability and biological realism. Wheat crops on Yorke Peninsula, South Australia, were studied to evaluate this approach. This paper, comprising part of the study, is a principal component analysis of the soil properties which form a natural subset of the variables measured. It shows that the variation of soils in the study area can be characterized by the variation of relatively few soil properties.