Multidimensional spatial processes are often encountered when soil variables such as temperature and water and salt content are observed as a function of space and time. Time- or spatial-domain methods based on state-space models are useful for describing the movement of these series through space and time. The detection of important periodicities can be accomplished using frequency-domain generalizations of regression and analysis of variance. These periodicities and the movement of these series through space or time reflect the nature of the soil-forming processes, agricultural practices (such as cultivation, irrigation, and fertilization) and the interdependency of these variables. Two experiments were conducted in the San Joaquin Valley of California to observe changes in soil properties over space as a result of different quality and quantity of irrigation water and to observe the cyclic behavior of soil properties as affected by quality of irrigation water and other agricultural practices. A number of spatial techniques were used to illustrate these changes.
© Williams & Wilkins 1989. All Rights Reserved.