Sampling and laboratory techniques
To measure soil properties correctly requires standard laboratory techniques and precise sampling methods. Historically different laboratory methods have been proposed to measure the same soil property leading to different and sometimes contradictory results (Lack on an agreement on the size of the clay fraction is one of the most annoying examples in this respect). Over the years most chemical methods have been refined and automated to such an extent that variability between results has been minimized. However, comparative analysis of results obtained amongst different laboratories still shows significant differences, sometimes larger than the change in the characteristic in time (for instance organic carbon, phosphorus).
Other soil characteristics show little variability of results between different laboratories (pH, and EC for instance), but may show significant differences depending on the time of the year the soil has been sampled. More recently a lot of interest and testing has gone in new automated methods such as spectroscopy that allows to measure a number of soil properties in a uniform way that is also cheaper than traditional laboratory methods.