Runoff Curve Numbers for Simulated Highway Slopes under Different Slope, Soil-Turf, and Rainfall Conditions

Highway infrastructure is a major non-point source of storm water runoff and associated pollutants. Forty-four simulated rainfall tests were conducted over a turf-covered plot-scale test bed that represented typical highway slopes of 25, 33, and 50%. The two soil-turf combinations were Argentine Bahia over AASHTO A-3 soil (fine sand) and Pensacola Bahia over AASHTO A-2-4 soil (silty fine sand). All the tests were conducted in central Florida, where the weather is subtropical. This paper presents the analysis of curve numbers (CN) obtained from simulated rainfall-runoff data. The CN values for the 24 tests conducted on the Argentine Bahia A-3 soil combination varied from 70.4 to 93.0 with a mean value of 79.6. The CN values for the 20 tests conducted on the Pensacola Bahia A-2-4 soil combination varied from 89.0 to 97.2 with a mean of 94.0. The scatter of the CN values was more for fine sand (standard deviation of 8.31) than for silty fine sand (standard deviation of 2.48). Single factor ANOVA suggested that the CN values are significantly lower (alpha = 0.05) for Argentine/A-3 compared with Pensacola/A-2-4. However, no significant difference (alpha = 0.05) was shown between the CN values calculated based on experimentally determined initial abstraction ratio (lambda) and the CN values calculated based on the Natural Resources Conservation Service (NRCS) suggested lambda of 0.2. The CN values, in general, increased with slope and rainfall intensity. The CN values determined in this study compare reasonably well with the 70-100 range of CN values reported in the literature for different types of turf-covered soil slopes. DOI: 10.1061/(ASCE)HE.1943-5584.0000605. (C) 2013 American Society of Civil Engineers