Influence of Initial Abstraction Ratios in NRCS-CN Model on Runoff Estimation of Permeable Brick Pavement Affected by Clogging

The widespread application of permeable pavements in urban stormwater management practices requires a quick and easy model to evaluate their hydrological function during different clogging stages. The Natural Resources Conservation Service-Curve Number (NRCS-CN) method has extensively used for predicting direct runoff from rainfall. However, the standard initial abstraction ratio (?) value of 0.2 in the model is the most ambiguous assumption and requires considerable refinement for the accurate runoff prediction of permeable pavements affected by clogging. This study calibrated the ? values of permeable brick pavement (PBP) with different clogging degrees to obtain an appropriate model to predict the runoff of PBP affected by clogging. The results indicated that the NRCS-CN model with standard ? value of 0.2 indeed overestimated the runoff over PBP surface, particularly at the initial clogging stage. The ? values showed a significant positive correlation with the clogging degree of PBP, while no significant correlation with the rainfall intensity. A function defining the parameter of ?, related to the saturated infiltration rate of PBP was developed. The accuracy of runoff prediction over the clogged PBP surface was greatly improved using the modified NRCS-CN model with the refined ? values. The NSE value rose to 0.96, and the MAE and RMSE values reduced to 2.66 and 3.85, respectively. Furthermore, the curve number (CN) values for the clogged PBP based on the calibrated ? values were much more consistent with the actual clogging rule of PBP, exhibiting an increased span and ranging from 35 to 98.