Does the uniform packing of sand in a cylinder provide a uniform penetration resistance? A method for screening plants for responses to soil mechanical impedance

Sand packed to a constant dry bulk density, is frequently used as an artificial growth medium in which to simulate the effects of a constant mechanical impedance on root growth. This research aimed to determine whether conventional packing resulted in constant mechanical impedance and to test alternative packing regimes. Perspex cylinders 300 mm tall with a 49 mm internal diameter were packed with moist sand to uniform and varying bulk densities to examine which type of packing gave the greatest uniformity of penetration resistance (PR) with depth. The cylinders packed to a constant bulk density (1.48, 1.55, and 1.6 Mg m(-3)) all had measured PR profiles which increased markedly with depth by approximately 1, 1.5 and 3 MPa, respectively, within the top 100 mm. Between 100-300 mm depth, these same cylinders showed reductions in PR of up to 1, 2 and 2 MPa respectively. These results show that sand packed to a constant bulk density with depth would not provide a uniform mechanical impedance to plant roots. By packing sand to different bulk densities at different depths, we obtained packed cylinders that had much more uniform PR profiles (with average values of 0.25, 1.40 and 2.30 MPa). Below a depth of 50 mm, the coefficients of variation for replicate cylinders packed in this way were 12%, 5% and 18% for the 0.25, 1.40, and 2.30 MPa treatments respectively. For experiments with single plants, the lower PR values that were unavoidable near to the surface (< 50 mm) can be avoided by sowing seeds at the base of a funnel inserted into the cylinder. Treatments such as these can provide reproducible growth media, with adequate water/nutrient and aeration status for the study of plant response to uniform mechanical impedance.