ABSTRACT
The best predictors of soil chemical behavior are mineralogy and specific surface. Mineralogy yields information on the physicochemical nature of surfaces, and specific surface provides information on the extent of the surface. Recent advances in quantitative mineralogy enable soil scientists to apply knowledge of mineral properties to predict soil behavior and performance. If, in addition to this, quantitative mineralogy is supplemented with total chemical analysis, the elements can be allocated to minerals and the amorphous fraction. Accessory properties that can be predicted from knowledge of mineralogy, specific surface, and elemental composition include surface charge characteristics, ion exchange capacity, noncoulombic adsorption-desorption reactions, aggregation, aggregate stability, pore size distribution, and water, energy, and gas transport coefficients. A classification system that includes mineralogy, including the amount and composition of the amorphous fraction, and specific surface as differentiating criteria should enable users to make better predictions of soil behavior and performance.
The goal of advancing soil chemistry and soil classification is to enable individuals who are responsible for the care and management of land to make better long-term predictions of how soils will behave and perform when used for specified purposes. Two ways to achieve this goal are to develop better prediction models and to improve the characterization of soil constituents. The latter is the more critical, because a model can only be as good as the knowledge we have about a soil’s characteristics.
While soil surveys and classification enable users to find useful information about how a soil will behave and perform, users are increasingly asked to make more precise predictions about the 68long-term consequences of land use practices. These include, for example, predictions of agronomic crop performance, the fate and whereabouts of agricultural and industrial chemicals in the vadose zone, and rheologic behavior of materials for engineering interpretation and use of soils. The irony is that soil characterization data are primarily used for soil survey and classification, and are rarely used as inputs for prediction models. Ideally, soil characterization data should be the same as input data for dynamic, process-based simulation models. If this were the case, it would result in greater demand and use of soil surveys and soil characterization data. But soil science is far from being ready for use as identical parameters for classifying soils and for use as input in prediction models. Soil texture and clay content are still the most commonly used predictors of soil behavior, but too often fail when used alone for this purpose. We know, for example, that a very fine Vertisol and very fine Oxisol will behave and perform differently. For this reason, the family category of Soil Taxonomy (Soil Survey Staff, 1999) specifies particle size as well as mineralogy for grouping soils into performance classes. But there are instances in which samples with nearly identical texture and mineralogy produce unexpectedly different results in the field and when analyzed in the laboratory for accessory properties, such as absorption isotherms, potentiometric titration curves, and zero points of charge.
The failure of particle size and mineralogy to predict soil behavior stems from the fact that both are surrogates of two other more fundamental soil properties, namely, specific surface and surface charge density. Although particle size substitutes for specific surface, we expect a very fine kaolinitic sample to have a lower surface area than a very fine montmorillonitic sample. Mineralogy also provides additional information about the sample’s surface charge characteristics. This implies that if specific surface and quantitative mineralogy were used as differentiating criteria in place of particle size and qualitative mineralogy, more consistent results would follow. I believe there is still some unfinished business in soil science that needs to be resolved before better prediction will be possible. This unfinished business is the quantitative analysis and characterization of noncrystalline or amorphous materials in soils.
