An Analysis of Terrain Classification for Long-Range Prediction of Conditions in Deserts

Bibliographic Data

ID4708071
AuthorsColin W Mitchell, R Webster (0000-0003-1767-7934), P H T Beckett (0009-0006-4414-1755), Barbara Clifford
Year1979
Volume145
Issue1
Pages72
Publication date1979-03-01
Peer ReviewedYes
Open AccessNo
TypeARTICLE
VenueGeographical Journal (JOURNAL)
Journal identifiersISSN: 0016-7398 • E-ISSN: 1475-4959
PublisherJSTOR (PUBLISHER)
DOI10.2307/633076
OpenAlexW2315544271
LanguageEN
Citations received2
References cited1

devised for classifying terrain, with the facet as the basic unit for practical land management, and coarser groupings to convey more general information for larger areas and to allow predictions about facets to be made over long distances. The hot desert terrain ofthe world was classified using it. This study examines the extent to which this classification achieved its purpose. Representative classes at all levels of the classification scheme were sampled, and several soil and other terrain properties measured at replicate sampling sites. Within-facet variances of each property were estimated for facets with scope ranging from the local form through each level of the hierarchy to the desert zone as a whole. Components of variance show how each level contributes to the total withinfacet variance. Within-class variances and components of variance are also pre? sented for the classes at all levels of the hierarchy. In general the largest contri? butions to the total variance derive from within-facet (residual), betweenfacet and between-province, in that order. Variances within local facets are always least, and for many properties can be small enough to allow useful prediction

Cartography · Geography · Physical geography · Range (aeronautics · Remote sensing · Terrain · Agriculture, Soil, Plant Science · Engineering · Environmental Science · Geology · Soil and Land Suitability Analysis

Tools

Open DOISci-Hub