Testing the pedometric evaluation of taxonomic units on soil taxonomy — A step in advancing towards a universal soil classification system

Abstract Most existing soil classification systems were developed to understand and provide information on soils, their natural properties and potential use for certain purposes. The conceptual developments of the systems took place before the recent boom of observation technologies, data storage and data processing achievements that can support to determine or predict soil differences. Until the recent past soils under agricultural or forestry use received more attention than other soils, such as anthropogenically modified or urban soils, or soils of the cold regions. The broader view of soil functions and the understanding of global environmental processes require a better understanding and description of all soils. Precisely recorded and harmonized data is needed to serve the new era of modern agricultural practices, other land uses as well as different scientific applications. The soil science community is challenged to apply the accumulated knowledge on soil formation, soil differences and functions, as well as, new tools of robust data processing to evaluate current systems and define objective relationships for better future classification systems. The evaluation of existing soil classification systems may help the understanding of taxonomic relationships of differentiated soil groups and improve our methods of classifying soils. This paper is summarizing the approaches and methods of evaluation that was applied for the great group and higher levels of Soil Taxonomy. Simple statistical and pedometric methods were applied on centroids and, calculated on the basis of properties commonly used to define the classification units. The centroids provide an objective tool to evaluate the concepts of taxa and the taxonomic relationships between them. Examples of conceptual evaluations and detailed discussions of the taxonomic distance calculations between the great groups within their orders and members of other orders are provided. The presented methods and relationships were found very useful for the evaluation purpose. The extension of the methods for other systems, other data bases and the combination of those is in progress. The initial results suggest that the objective, pedometric approaches can support the development of an envisioned Universal Soil Classification System.