Soil erosion under simulated rainfall in relation to phenological stages of soybeans and tillage methods in Lages, SC, Brazil

Soil tillage may increase vulnerability to water erosion, whereas no tillage and other conservation cultivation techniques are viewed as strategies to control soil erosion. The objective of this research was to quantify runoff and soil losses by water erosion under different soil tillage systems at the Santa Catarina Highlands, southern Brazil. A field study was carried out using a rotating-boom rainfall simulator with 64 mm h(-1) rainfall intensity on a Typic Hapludox, between April 2003 and May 2004. Five rainfall tests were applied along successive cropstages. Surface cover was none (fallow) or soybean (Glycine max, L). Five treatments were investigated, replicated twice. These treatments were conventional tillage on bare soil (BS) as a control treatment and the following treatments under soybean: conventional tillage (CT), no tillage over burnt crop residues on never before cultivated land (NT-B), no tillage over desiccated crop residues, also on never before cultivated land (NT-D) and traditional no tillage over desiccated crop residues on a soil tilled 4 years before this experiment (NT-PT). Water losses by surface runoff seemed to be more influenced by vegetative crop stadium than by tillage system and consequently a wide range of variation in surface runoff was found, following successive cropstages. The most efficient tillage system in reducing surface runoff and soil losses was no tillage, particularly the NT-PT treatment. Sediment losses were more influenced by tillage system than water losses. In the NT-B, NT-D and NT-PT treatments the rate of sediment losses along the crop vegetative cycle showed a tendency to increase from the first to the second cropstages and later to decrease from the third cropstage onwards. In the conventionally tilled treatment (CT) soil losses were greater than in any of the no tillage treatments (NT-D, NT-B and NT-PT) during the initial growth periods, but at the end of the vegetative period differences in sediment rates between tilled and non-tilled treatments tended to be smaller. In the BS control treatment, soil losses progressively increased following the vegetative growth season of soybean. (C) 2008 Elsevier B.V. All rights reserved.