Date of this Version
Imagine that you are entering an unknown field and would like to estimate the productivity of the unfamiliar soil. You may pick up a handful of soil to evaluate its color and texture. You also can feel how difficult it is to break a clod apart, roll it into a ball or press out a ribbon. After repeating this procedure at different field locations, soil depths and times, you get a feeling of both spatial and temporal soil variability. Some of this variability can explain the non-uniformity of crop yield. If you collect soil samples and send them to a soil-testing laboratory, you can get a standardized measure of soil nutrient levels and other characteristics. The greater the sampling density, the more likely you are to obtain a good representation of the variability of soil properties across the field. This process, however, takes time and money, both when sampling and in the lab, and limits the number of soil samples which can be justified economically.