TILLAGE, ROW SPACING, AND CULTIVATION AFFECT EROSION FROM SOYBEAN CROPLAND

Soybean plots were planted in a replicated 2 x 2 factorial arrangement of row width [0.18-m (7-in.) or 0.91-m (36-in.) rows] and tillage [wi thout prior tillage (no-till) or following primary and secondary tilla ge]. Additional unreplicated conventional and no-till plots were plant ed in 0.91-m rows and were cultivated after planting. A rainfall simul ator was used to apply a two-rainstorm sequence to a 9-m section of ea ch plot when the soybean plants had from 6 to 11 main stem nodes (four to eight weeks after planting). Inflow was added during the second ra instorm to simulate longer slope lengths. Sediment size distribution a nd rill development were measured. The study was conducted in 1989 (th e first year of no-till, after one year conventional tillage, after so d) and repeated in 1991 (second year no-fill after treatments were rer andomized on plots in 1990). Results were similar for both years. In g eneral, erosion rates were several times higher with conventional till age than with no-till planting at the crop stage tested Planting in 0. 18-m rows doubled erosion rates compared to wide row planting during this crop growth stage. However, most striking was the large increase in erosion caused by row cultivation. Cultivation without intervening rainfall prior to the simulated rainstorm sequence increased soil loss 20-fold for the conventional tillage system and 40-fold for the no-ti ll planted system. Erosion rates from a 70-mm rainstorm in one hour on a simulated 60-m, 6% slope planted in 0.91-m rows oriented up-and-dow n hill and cultivated averaged 57 Mg/ha for conventional tillage and 3 6 Mg/ha for no-till planting. Although the silt loam soil contained es sentially no sand, between 20 and 30% of this sediment eroded as sand- sized aggregates.