Mitigation of subsoil recompaction by light traffic and on-land ploughing. II: Root and yield response

Plough pans have been shown to severely hamper root development, limit rooting depth and reduce crop yields. We evaluated the effect of plough pan compaction on root and yield response for a winter wheat in a field trial conducted in two neighbouring fields (B3 and B4) on a sandy loam. Plots were mechanically loosened to a depth of 35 cm in 1997 (B3 and B4) and again in 1998 (only B4). Perennial grass/clover was grown with limited traffic intensity in 1998-1999 and 1999-2000 for B3 and B4, respectively. The perennial grass/clover was ploughed under in spring 2000 (B3) and spring 2001 (B4) and oats established. After harvest of oats, winter wheat was established. Recompaction treatments were applied to the mechanically loosened plots in 2000 and 2000+2001 for B3 and B4, respectively. On-land ploughing was compared with traditional mouldboard ploughing with the tractor wheels in the furrow. In addition, the loosened plots were either heavy trafficked (10-18 Mg axle load and ~200 kPa inflation pressure) or light trafficked (<6 Mg axle load and <100 kPa inflation pressure). The loosened treatments were referenced by plots where the plough pan had not been mechanically loosened (CONV). Root growth of winter wheat was followed during both growing seasons applying the minirhizotron technique. In 2002 these measurements were supplemented with core sampling at around anthesis (June 11). Soil water content was followed during the 2002 growing season using time domain reflectometry (TDR). Grain yield and nitrogen content in grain were determined both years. An adjoining study showed that the combination of heavy traffic and traditional ploughing (NINV-TH) caused strong recompaction of loosened soil whereas the combination of light traffic and on-land ploughing produced moderate recompaction. This study showed that the NINV-TH treatment produced 7% lower yield than in the NINV-OL treatment in 2002, whereas no clear effect on crop yield was found in 2001. The NINV-TH and NINV-OL treatments showed mixed effect on root growth. Surprisingly, the non-loosened CONV treatment performed similar or even better than the NINV-OL treatment. The CONV treatment facilitated higher root intensity at depth and produced similar yield and N-uptake. The good performance of the CONV soil may be related to detrimental effects of subsoil loosening on rootability, favourable water supply in both experimental seasons, and improvements of the CONV plough pan structure through biological tillage during the experimental years. Our results suggest that mechanical subsoil loosening of humid sandy loams only is recommendable in case of very severe subsoil compaction. Natural alleviation of subsoil structure induced by changes in soil management may comprise a favourable alternative to mechanical subsoil loosening.