Evaluating the Accuracy of Common Runoff Estimation Methods for New Impervious Hot-Mix Asphalt

Abstract

Accurately predicting runoff volume from impervious surfaces for water quality design events is an important step to meet water quality and infiltration design targets for green infrastructure stormwater control measures. The objectives of this research were to quantify abstraction from a recently paved impervious hot-mix asphalt (HMA) parking lot surface and evaluate the accuracy in modeling runoff volume from small events using the following runoff estimation methods: the Soil Conservation Service Curve Number (SCS-CN) method, the Simple Method (SM), and the Small Storm Hydrology Method (SSHM). A 4,000 m2 (0.4-ha) parking lot in Edison, New Jersey, was designed with impervious HMA driving lanes evenly draining onto parking lanes constructed with permeable pavement. Ten lined permeable pavement sections that capture all infiltrating water and route it to collection tanks were included in this study. Using a water balance approach on an event basis, the measured infiltrate volume was compared to the rainfall volume across the drainage area to determine the rainfall retained by the HMA surface and in the permeable pavement strata and underlying aggregate. Only events with an antecedent dry period (ADP) less than 24 h (N=16) were used in this analysis because it minimized evaporation and additional storage in the permeable pavement profile. It was assumed that the rainfall retention depth for these events was completely abstracted in the HMA surface. In comparing measured retention in the HMA surface to the three runoff estimation methods, the SCS-CN method overpredicted abstraction in half of the test sections, and the SM and SSHM underpredicted abstraction in 8 of the 10 test sections. While evaporation from the permeable pavement profile from the events analyzed was small, it was not zero, so the measured rainfall retention depth had a small positive bias. After correcting for this bias, the results shifted closer to the predictions using the SM and SSHM and farther from the predictions using the SCS-CN method. The average and interquartile range (IQR) of the corrected depression storage depth in the HMA surface for the 10 test sections were 2.0 mm and 0.9–3.0 mm, respectively. For a 25.4-mm water quality design event, the predicted abstraction depth by the SM was within the IQR, the SSHM prediction was 0.1 mm below the IQR, and the SCS-CN prediction exceeded the IQR by 2.3 mm. The runoff predicted by the SCS-CN method for this example water quality event was about 15% less than the average and about 10% less than the IQR bound.