Non-inverting soil tillage as a means of optimising soil tilth

A field experiment was initiated in 1997 with non-inverting soil tillage compared to conventional tillage with annual mouldboard ploughing to either 20 cm depth (CT) or 10 cm depth (RCT). The new system included a non-inverting mechanical loosening of the soil to approximately 35 cm depth combined with a shallow cultivation of the top 5 cm soil by either a rotovator (NIT) or an S-tine harrow (RNIT). The experiment was located on a sandy loam at the organic farmed Rugballegård Experimental Station. A plough pan at 20-35 cm depth was identified by cone penetration measurements. The non-inverting tillage system succeeded in breaking up the plough pan, which reduced the cone index from approximately 1.8 MPa in the CT and RCT treatments to 0.6 MPa in the NIT treatment. A detailed soil profile description of the top 30 cm of the soil indicated that root growth was restricted for especially CT and RCT treatments in the form of thickened and horizontally deflected roots at the interface between the topsoil and the plough pan at approximately 22 cm depth. For RCT a new plough pan was under development just below the present ploughing depth at 10 cm. This also resulted in root growth restriction, but to a lesser extent than the deeper plough pan. A soil drop test was performed to characterise soil fragmentation. Undisturbed cubic soil samples were taken from the 5-15 cm layer and dropped from 75 cm height in to a metal box. The resulting aggregate size distribution differed between the treatments with geometrical mean weight diameter (GMD) of 5.7 mm 7.4 mm and 8.9 mm for CT, RCT and NIT treatments, respectively. The higher degree of fragmentation for the mouldboard ploughed soil may be explained by higher energy input per soil volume by mouldboard ploughing than by non-inverting deep loosening. The results indicate for the 5_15 cm layer, that plant growth conditions were best for CT treated soil, due to a combination of low cone index and small aggregates. For the 20_35 cm layer, the results indicates best plant growth conditions for NIT treated soil due to a low cone index and less restricted root growth. The spring barley/pea mix with grass/clover undersown yielded on average 47.6 hkg/ha. There were no significant differences between treatments.