The objective of this study was to evaluate the effects of cover crops grown under no-tillage on the aggregation and physical protection of organic matter in soil macro and microaggregates. The experiment consisted of a randomized complete block design with three replications. The following cover crops were investigated in corn rotation systems: T1 = Braquiária ruziziensis (Urochloa ruziziensis), T2 = Canavalia brasiliensis Mart. ex Benth.,Br.], T5 = turnip-forage (Raphanus sativus L.), T6 = velvet bean (Mucuna aterrima Merr.) and T7 = native Cerrado vegetation as a reference environment. Soil was sampled at a depth of 0-10 cm in September 2015 for the determination of organic matter fractions in macro and microaggregates. There was a reduction in aggregate size and its stability when native Cerrado areas were converted into agricultural systems. Nevertheless, some cover crops such as velvet bean, millet and turnip-forage favored restructuring the soil, forming stable aggregates similar to the native Cerrado. Among the cover crops, millet was highlighted as presenting elevated capacity to accumulate labile organic carbon in macroaggregates (2.32 g C kg -1 ) and microaggregates (2.34 g C kg -1 ). These values are, on average, 60% higher than those presented by turnip-forage. In general, the conversion of land use under Cerrado vegetation to agroecosystems reduced the total organic carbon content, mainly due to macroaggregate breakup, resulting in a lower physical protection of soil organic matter.

The use of intensive mechanization in no-tillage areas can change soil physical conditions, mainly in the "Cerrado", which has reduced cover ratio. This study aimed to evaluate physical and mechanical properties of a red latosol under no-tillage and subjected to simulated dynamic traffic in different surfaces conditions. For this, soil samples were collected, with dimensions of 0.2 × 0.2 × 0.3 m (height, width and length), in an area subjected to no-tillage in the last four years. Subsequently, samples were transported to the laboratory and subjected to different traffic levels in a simulator. Shortly after, a completely randomized experimental design was set up, in factorial 2 × 5, with two covers (with and without haystack) and five traffic levels (zero, one, two, four and eight run overs) applied by the simulator. Assessed physical properties were superficial settlement, soil density, compaction degree and pre-consolidation pressure. Results showed that superficial settlement, soil density and compaction degree were significantly influenced by soil cover and traffic intensity. Pre-consolidation was not affected by cover, and had higher values when subjected to more traffic intensity.