Premium
Defining Ranges of Soil Characteristics
Author(s) -
Jansen I. J.,
Arnold R. W.
Publication year - 1976
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1976.03615995004000010025x
Subject(s) - range (aeronautics) , sample (material) , property (philosophy) , population , mathematics , scale (ratio) , distribution (mathematics) , statistics , soil science , interval (graph theory) , environmental science , geography , cartography , mathematical analysis , physics , combinatorics , engineering , philosophy , demography , epistemology , sociology , thermodynamics , aerospace engineering
A method is developed for using quantitative sample data to characterize mappable bodies of soil. Inclusions are separated into two kinds. The first consists of included bodies of soil that are recognized as separate entities, but not mappable because of scale limitations. The second consists of scattered areas having extreme expression of some significant property but not recognized as separate geographic entities. Inclusions of the first kind are sampled separately and treated as separate entities. The location and setting of these inclusions can be described in relation to the dominant soil body. A range of expression for each measurable soil property of interest is determined separately for each soil entity. The ranges are derived from the sample data and are designed to include a fixed portion of the total population of values rather than to be all inclusive. Soil areas having extreme expression of some particular property (falling outside of the defined range) are inclusions of the second kind. The location and setting of those soil areas cannot be readily described but their extent is controlled. The tolerance interval procedure is used to calculate range limits from sample data when a normal distribution can be assumed. The resulting range can be said to include a predetermined portion of the total population of values with a specified degree of confidence. When the distribution is believed to deviate markedly from normal, the range limits are set subjectively by observing a frequency distribution of the sample data. Ranges set by the tolerance interval procedure were generally somewhat wider than those set subjectively, because the tolerance interval procedure allows for the imperfect relationship between the sample and the population and enables one to make confidence statements about the ranges defined.