Impacts of distinct spatial arrangements of impervious surfaces on runoff and sediment fluxes from laboratory experiments

Abstract Urbanization affects runoff processes and sediment transport, but the magnitude of the impacts remains poorly understood. Different spatial patterns of pervious and impervious surfaces influence flow and sediment connectivity between hillslopes and stream networks. Following years of research on the peri-urbanizing Ribeira dos Covoes catchment in Portugal, this study uses laboratory rainfall simulation experiments to better assess the impact of  soil and pavement patterns on runoff (amount, runoff start and stop times) and sediment transport. Based on urban cores observed in the study catchment, the investigation focused on seven spatial patterns: bare soil (S), 100% pavement (P), and 60% pavement under continuous - C - surface placed upslope (CU) and downslope (CD), and dispersed - D - over the surface with regular (DR), irregular (DI) and linear (DL) distribution. A 1.00 m × 1.00 m flume, 0.05 m deep with a 9° slope, facilitated the experiments. The study used sandy-loam soil (1500 kg m-3) with concrete slabs representing pavement. Each experiment comprised a series of four rainfall simulations, each lasting 20 minutes (50 mm h-1), separated by 30-minute intervals,  to assess the impact of different initial soil moisture conditions. Results indicate that both spatial pattern and soil moisture drive runoff. Under dry conditions, CD provides runoff that is 7 times faster and about 4% higher than that for CU. Already wet conditions, however, produced 12% more runoff on CU than on CD. The greater runoff arose from faster soil saturation, driven by soil moisture increasing more quickly during the rainfall, as well as upstream runoff from paved surfaces, though runoff took longer to reach the outlet. The dispersed pavement pattern only affected runoff amount, with DL producing the highest coefficients (40-71%) and DI the lowest (25-55%), since longer flow paths increase the opportunities for water infiltration.  Additionally, CU yielded 40% more sediment transport  than CD, but the three dispersed patterns did not show a significant impact (p>0.05). The results suggest that appropriate planning can reduce flood hazard and land degradation in urban areas, in particular by using dispersed patterns of sealed surfaces to enhance water infiltration and retention.