Erosion–productivity–soil amendment relationships for wheat over 16 years

Soil erosion remains a threat to our global soil resource. This study was conducted to ascertain the effects of simulated erosion on soil productivity and methods for its amendment. Incremental depths (0, 5, 10, 15, and 20 cm) of surface soil, or cuts, were mechanically removed to simulate erosion at two sites (one Dryland, one Irrigated) in southern Alberta in 1990. Three amendment treatments (nitrogen + phosphorus fertilizer, 5 cm of topsoil, or 75 Mg ha−1 of feedlot manure) and a check were superimposed on each of the cuts. The sites were cropped annually until 2006. Average grain yield reductions during 16 years were 10.0% for 5 cm, 19.5% for 10 cm, 29.0% for 15 cm, and 38.5% for 20 cm of topsoil removal. There was evidence that the restoration of productivity levelled off at a value less than the non-eroded treatment rather than gradually converging on it, within the timeline of the study. Average grain yield loss was 50 kg ha−1 cm−1 yr−1 at the Dryland site and 59 kg ha−1 cm−1 yr−1 at the Irrigated site. As the depth of cut increased, the average residual effect (1993–2006) of manure increased, e.g., on the 5 cm cut, the residual effect (over the equivalent cut check treatment) was 20.9%, climbing to 41.9% on the 20 cm cut. Amendments ranked manure > topsoil > fertilizer in terms of restoring productivity to the desurfaced soils. The average grain yield during 16 years on the check treatment fell 2.1% cm−1 depth of topsoil removal on the Dryland site and 1.7% cm−1 for the Irrigated site. In contrast, grain yield on the manure treatment fell by 0.8% cm−1 on the Dryland site and 0.9% cm−1 on the Irrigated site. The study reinforces the need to prevent erosion and indicates that application of livestock manure is an option for restoring soil productivity in the short term.