Scientific Systems
Scientific systems are of two types, those based on processes of soil development or genesis and those based on quantifiable characteristics. Many systems have elements of both types. Criteria for distinguishing soils among these systems are by no means uniform. The structure of these systems can be hierarchical, descriptive, or nominal. Also, soil classification systems are not static. As knowledge is gained old systems and class names are changed for new ones, as with plant and animal classification systems. These differences in classification systems make it important to include descriptions of classified soils when reporting so that correlation to other systems is possible. Users of classification systems need to make sure that important characteristics have class limits narrow enough to be useful. Scientific systems, especially the hierarchical ones, are useful to exclude soils from consideration, but generally a detailed soil description is needed to make recommendations for soil use and management.
Process based classification systems were developed to explain how soil characteristics and appearance change with time. Soils form distinct layers as biological, physical, and chemical processes develop zones of material accumulation and zones of loss, also called horizons. This evolution is predicable as long as climate and other processes remain constant. This assumption of constancy poses problems for these types of classification systems, especially for those aridic soils that partially developed under humid conditions. It is possible to have more than one classification for the same soil, depending on which soil forming process is assumed to be expressing itself. Due to this and other problems, the USDA abandoned its system based on soil forming processes and developed the system layed out in Soil Taxonomy based on existing quantifiable characteristics. Engineers also use systems based on existing characteristics, for obvious reasons.
Criteria for distinguishing soils are not necessarily the same for different classification systems. Soil texture is a good example of the problems involved in correlating classifications between two or more systems. Texture describes the proportion of different size classes of the mineral part of the soil. For the same soil, texture-class names can differ depending on the classification system used. Even if the names are the same, the limits often differ, as with "sandy clay" of the USDA and French systems:
The problem continues with particle-size classes. For example, clay is defined as <0.002 mm in diameter by some systems and <0.005 mm by others. Other physical, chemical, and biological characteristics have similar discrepancies between systems which makes one to one correlation between systems nearly impossible unless a detailed soil description is available.
Hierarchical, descriptive, or nominal systems of classification each have their strengths and weaknesses. Rules of hierarchical systems allow one to determine the highest to lowest levels of classification without previously knowing any of the possible classifications. The resulting classification (e.g., clayey-skeletal, mixed, hyperthermic Typic Haplargid of USDA Soil Taxonomy), while providing much information, is far from what is usually needed to make management decisions. Also, these systems can make trivial distinctions when applied in areas of the world different from those where they were developed. For example, sandy West African soils have extremely low cation exchange capacities. Very small additions or losses of base cations can cause large differences in base saturation percentages that distinguish the orders Alfisols and Ultisols (highest level of USDA Soil Taxonomy).
Descriptive soil classification systems are commonly developed for single purpose application. For example, the Fertility Capability Soil Classification System uses a string of upper and lower case letters to represent characteristics important in soil fertility management for crops. These systems are relatively simple, easy to interpret, but of limited value.
Nominal systems are much like vernacular systems but with the rigor of scientific descriptions and engineering capabilities. These systems can be detailed enough to use as a basis for management decisions, but only at a regional level. Comparing numerous soils under this system would be an onerous task. The USDA's system of Soil Series is an example of a nominal system that predates Soil Taxonomy. Soil Series have been modified so that they fit within one Soil Taxonomy classification and linked to this hierarchical system. Together, Soil Taxonomy and Soil Series systems provide powerful tools for identifying, understanding and managing soils. This merging of hierarchical and nominal systems can be copied in other countries, especially in developing countries by using existing vernacular systems.
To demonstrate how this all fits together, the aridic soil "Malibu blue clay" (vernacular name) was established by the USDA in 1910 as Diablo soil series (old nominal scientific system) and classified as a Grumusols (old hierarchical scientific system). Since then the Diablo soil series description has been modified (new nominal scientific system) and classified as fine, smectitic, thermic family of Aridic Haploxererts (current hierarchical scientific system).
(Source: http://ag.arizona.edu/OALS/IALC/soils/classifsystems.html)