Towards a technique for determining basal cover in tufted grasslands

Abstract

Computer simulation, point-based field sampling procedures, and step-wise multiple linear regression were used to establish a relationship between the basal cover (BC) of a sample site and paired observations of the mean distance from a point to the nearest living tuft (D)and the mean basal diameter of the tufts (d). The model may be summarised as: BC = 19.8+0.39(D)-11.87(loged)+0.64(d)+2.93(loged) Repeatability of the distance and diameter measures was high. Predicted basal cover derived from the equation proved to be more sensitive, but less precise, at low distance values than at high distance values. Whilst a number of shortcomings of the approach are highlighted it is suggested that the technique will provide an effective means of indexing the basal cover of tufted grasslands.

Keywords: basal cover; botany; computer simulation; distance measures; equation; grasslands; model; monitoring; precision; procedures; regression; repeatability; sampling; south africa; technique; tufts

African Journal of Range & Forage Science, 10(2): pp. 77-81