|
|
||||||||
ABSTRACT
Soil classification systems attempt to group soils that have similar properties. One test of how well a soil classification system does this is to classify soils having wide ranges of important soil properties and then determine how much of the variation in soil properties is partitioned among the taxonomic categories and how much variation remains within a category. Eighty soil samples (Andepts) from the Island of Hawaii were collected and analyzed. Phosphorus, pH, exchangeable cations, and silicon were determined. The ability of Soil Taxonomy to predict variation in these soil properties was examined in two ways. The first approach measured how much of the variation in soil properties was partitioned by single categories of the Soil Taxonomy. In the second approach, we prepared a hierarchical analysis of variance patterned after the hierarchical classification sequence actually used in Soil Taxonomy. Both methods indicated that most of the variation in soil chemical properties was predicted or partitioned by the separation of the 80 soil samples into four great groups. The soil series accounted for the second largest amount of variation in soil chemical properties. The other categories contributed relatively little to the explanation of variation. This appears to be desirable because it permits a new set of differentiate to be incorporated at each taxonomic category. The results from the hierarchical analysis of variance and proportioning the variation within taxonomic categories suggested that correlations among the differentiae may alter the partitioning of variation from that which should occur with independent differentiae. Separation of Hydrandepts from the other three great groups accounted for 20 to 50% of the variation in those soil chemical properties measured that could be explained by classifying the soils to the mapping unit level. Ninety-two percent of the variation in rainfall was partitioned by classifying the soils to the series level.
1 Journal Series no. 2520 of Hawaii Agric. Exp. Stn. This work was supported by a 211(d) Basic Grant (AID/CSD2833). All reported opinions, conclusions, and recommendations are those of the authors and not those of the funding agency or the U.S. government.
2 Assistant Soil Scientist and Professor of Agronomy & Soil Science, respectively, Dep. of Agronomy & Soil Science, Univ. of Hawaii, Honolulu, HI 96822.
Received for publication May 30, 1980. Accepted for publication November 7, 1980.
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||