THE RELATION BETWEEN COST AND UTILITY IN SOIL SURVEY (I–III)1

Publisher Website
Google Scholar
Abstract
Summary: A series of five papers compares the cost‐effectiveness of different procedures for soil survey at medium scale. The first three are presented here.The whole trial area of 120 km2 in Berkshire, in south‐central England, was mapped in soil series by free survey at 1:25 000 for publication at 1:63 360. Three contrasting sample areas of 1.26 km2 were resurveyed to the same legend by free survey at 1:10 560, and by grid survey at a range of scales between 1:20 000 and 1:70 000, to both general purpose (soil series) and fifteen–twenty different single‐property legends. The direct costs of producing each map were recorded.The study confirmed how much the free survey procedure depended on the external features of soil boundaries to locate them. The density of soil observations required to map soil series by free survey at the same map scale in different terrains was approximately proportional to the length of mapped boundary/km2, or to the number of separately mapped soil occurrences/km2. The density was least where the soil boundaries had the clearest external expression. Survey effort/km2 increased in proportion to the density of observations but was also affected by local differences in the ease of cross‐country access, or in the effort necessary (by spade or auger) to identify the soil at a point.For soil series grid surveys there are approximately linear relations between log(cost) and log(map scale) with gradients between 1.3 and 1.7.A map of soil series by grid survey is more expensive than a map of the same units, based upon the same density of observations by the same surveyor, by free survey. But a series map by grid survey by a scientific assistant is cheaper than a series map based on the same density of soil observations by free survey by a scientific officer (diplomate or graduate). The cost of an isoline map of a single soil property depends very much upon the cost of determining the property mapped, and to some extent upon the number of different isoline maps produced from a single set of samples or observations. Even at the unusually low costs of chemical analyses assumed here, an isoline map of one chemical property costs nearly twice as much as a series map by grid survey.

This publication has 12 references indexed in Scilit: