ARS | Publication request: RUNOFF AND SOIL LOSS FROM MIDWESTERN AND SOUTHEASTERN U.S. SILT LOESS SOILS AS AFFECTED BYTILLAGE PRACTICE AND SOIL ORGANIC MATTER CONTENT

Submitted to: Soil & Tillage Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: January 29, 2002
Publication Date: June 1, 2002

Interpretive Summary: The reductions in runoff and soil loss following the introduction of no- till practices are generally attributable to increases in organic matter contents at the soil surface, but the mechanisms involved have not been evaluated at a wide range of organic matter contents that have accumulated over long periods of time. Our results showed that no-till reduced runoff from cotton by 17% and 26% from corn after 9 years, compared to conventional tillage. After 35 years, runoff from no-till corn had decreased nearly 80%. Soil loss was reduced by 100% in all cases. These reductions in runoff and soil loss are the result of greater organic matter contents which increased aggregate stability, and lowered water dispersible clay contents. These results can be used in erosion models to predict changes in runoff as a function of time following the introduction of no-till practices.

Technical Abstract: The introduction of no-till practices increases soil organic matter (SOM) contents to the extent that reductions are generally observed in runoff (RO) and soil loss (SL). This research was conducted to determine the effects of SOM contents on RO and SL from two soils using crops that produced different amounts of residue. Rainfall simulator plots were installed on a 9 yr old tillage study at Senatobia, MS, and a 35 yr old study at Coshocton, OH. All RO was collected from two replications of conventional (CT) and no-till (NT) treatments imposed on corn (Zea mays L.) and cotton (Gossypium hirsutum L.) yield plots. Soil samples collected to a depth of 15 cm were characterized for SOM contents, aggregate stability (AS), water dispersible clay (WDC), and particle size distributions. Additional samples were collected for bulk density (BD). Runoff from the CT and NT treatments averaged 275.6 and 164.3 Mg ha**-1, respectively. Soil loss from the CT treatments averaged 3.9 Mg ha**-1. No measurable SL was recorded for the NT treatments. The SOM contents, AS, and WDC contents averaged over the surface 3 cm were 18.2 g kg**-1, 27.2%, and 7.9% for CT, and 53.6 g kg**-1, 63.3%, and 8.6% for NT. Bulk densities in the surface 3.8 cm averaged 1.34 Mg m**-3 for CT and 1.26 Mg m**-3 for NT. Regression models indicated that AS, WDC, and BD explained 97% of the variability in RO from the CT treatments, and AS as a single variable accounted for 83% of the variability in RO from the NT treatments. These results indicate that as SOM contents gradually increase in NT treatments, RO can be expected to decrease due to improvements in AS which reduce surface sealing, increase infiltration,