A comparison of four methods for measuring roots of field crops in three contrasting soils

Abstract

Rooting measurements have been made at different growth stages for sugar beets (1987) and for cereals (1988) on three different sites using four different root measurement techniques: (a) the core method where roots were extracted and root length is directly measured, (b) the core-break method where the visible roots were counted on the faces of a broken soil column, (c) the trench profile wall method where the number of visible roots were counted and the root length density was estimated on a profile wall, and (d) the monolith method where the roots were extracted from monoliths dug out from a profile wall. The calibration curves between the field methods and the extraction methods were not linear, and regression coefficients differed significantly between different sites, crops and between fields with different agronomic management, e.g. irrigation and liquid manure application. Differences between growth stages were comparably low compared with those found between locations. Root length densities obtained with the trench profile method were on average 10-fold lower in the sand brown earth, 6-fold lower in the vertisol and 4 times lower in the cambisol compared to data obtained with the core method. It is therefore concluded that the core-break method and the trench profile wall method deliver no reliable data for comparing rooting intensities between different soils and between different crops if they are not calibrated with an extraction method for each site and crop.