Effect of conventional and minimum tillage on physical and biochemical stabilization of soil organic matter

The objectives were to investigate (1) to which extent water-stable macro- and microaggregates sequester organic matter (OM) in a minimum tillage (MT) system compared to a conventional tillage (CT) system and (2) if the content of biochemically stabilized OM differs between both tillage systems, and (3) to study the temporal dynamics of the distribution of aggregate size classes and of storage of OM within aggregates in the field. Surface soils (0-5 cm) and subsoils (10-20 cm) were sampled after fallow (March 2007) and directly after tillage (November 2007) from a long-term experimental field near Gottingen, Germany. Macroaggregates (> 0.25 mm) were in general less abundant after fallow than directly after tillage. In March, only 21% (CT) and 45% (MT) of C(org) was stored within macroaggregates in the surface soil, whereas in November, the percentages increased to 58% and 73%, respectively. CT and MT soils of both depths were incubated as bulk soil (CT(bulk), MT(bulk)) and with macroaggregates disrupted (< 0.25 mm) (CT(md), MT(md)) for 28 days at 22A degrees C and water content of 50% of the maximum water holding capacity. For the MT(bulk) and MT(md) surface soils, C mineralization was significantly higher compared to the CT soils. Incubation of md soils did not generally result in a significantly higher C mineralization compared to the respective bulk soils, except for the MT(md) subsoil. Acid hydrolysis showed that the proportion of biochemically stabilized, nonhydrolysable, C(org) to total C(org) was lower in the MT than in the CT soils. Overall, the data indicate that the effect of physical stabilization of OM stored in the macroaggregates may not be a mechanism protecting very labile C with a turnover time of weeks, but that longer preservation likely occurs after macroaggregate transformation into microaggregates, and the surplus of OM found in the surface soil of MT does not only depend on the biochemically stabilized OM. Finally, our data suggest that the temporal variability of distribution of aggregate size classes in the field is large, but spatial and operator variability also contributed to the observed differences.