QUALITY CONTROL IN SOIL SURVEY

The paper examines 66 Australian soil surveys in a variety of terrains (but not close forest), by several survey procedures, and published at a range of map scales. It relates the Survey Effort (E) of professional staff (in man-days per km²) to (I) survey procedure, (2) the kind of mapping unit, and (3) the intricacy of the soil pattern mapped. Intricacy (I), the average number of mapped soil boundaries crossed by 1 km of random linear traverse, is related to the total length of mapped boundary (km per km²). When the surveys are grouped according to survey procedure and mapping unit, the survey effort for each group may be described by a regression of the form . B could not be shown to differ significantly between groups. D varied in the ratio 0.5: I, according to whether or not surveys used air photograph interpretation, and in the ratio 0.3:0.7:1.o, according to whether they mapped land systems, other compound mapping units, or simple mapping units. Since the choice of survey procedure and mapping unit is usually governed by the intricacy of the soil pattern the effect of these factors can be summarized in a single regression for all 66 surveys: There is a significant (P≤ 0.001) log-log regression between I and map scale.