Slope length effects on runoff and sediment delivery

ABSTRACT:

Lab studies provide an opportunity to isolate processes influencing water, sediment, and agrichemical transport under standard conditions. However, extending this information to field or watershed scales is often difficult. We compared runoff (R) and sediment (E) losses from a lab study with field data from a Tifton loamy sand (3% slope). Three plot scales/length-rainfall simulator methodologies were used-1.) 0.32 m² lab pan (L = 0.6m) under an oscillating nozzle rainfall simulator; 2.) 5.5 m² field plot (L = 3 m) under a Wobbler nozzle rainulator, and 3.) 600 m² field plots (L = 43 m) under the same rainulator used in method 2. For field plots (methods 2 and 3), R and E losses were measured from six simulated rainfall events (I = 25.4 mm hr⁻¹, 2 hr duration) during two corn growing seasons (five days before agrichemical application and 1, 14, 23, 43, and 108 days after agrichemical application). Similar rainfall intensity and duration were used in the lab study. R and E losses from lab pan (method 1) and 5.5 m² field plots (method 2) were measured at 5 minute intervals, whereas R and E delivery for 600 m² plots (method 3) were measured continuously and at selected times, respectively. R and E rates from all method generally increased during each event with similar maximum rates. Total R and E for method 2 was at least an order of magnitude greater than those for method 1, and total R and E for method 3 were at least 1 order of magnitude greater than those for method 2. R and E were related to slope length (R² = 0.94–0.99). Exponents (b) for R and E were 0.50–1.63. Detachment and transport processes varied spatially. Once a critical slope length was exceeded, rilling occurred Rilling was non existent in method 1 and was present but not dominant in method 2. For method 3, slope length was sufficient to cause rilling, therefore E was greater than that for methods 1 and 2.