A statistical method for obtaining an hour-level design storm of urban stormwater source control.

In urban stormwater source control design, the current method statistically obtains the design target-daily rainfall relationship and uses daily rainfall as the design storm to calculate the size of source control. Since it lacks the rainfall duration needed for sizing calculations under the design storm, it cannot ensure that the design solution achieves the design target. To overcome this problem, this study proposes a statistical method to obtain an hour-level design storm. It includes hour-level event identification, the concept of design storm event set, and a method to obtain the design target-rainfall-duration-maximum 1-h rainfall relationship. Model-based case study results suggested that rainfall volume control was more suitable than rainfall event percentile control as a design target because design solutions achieved the former but not the latter. Across six climate conditions, rainfalls and durations had logarithmic correlations ranging from 0.8454 to 0.9868. The positive near-monotonic relationships supported the source control sizing calculation under the design storm. Besides, a maximum 1-h rainfall could be used to calculate the hydraulic conductivity of planting soil layers, ensuring that the runoff peak penetrates the source control rather than off-site discharge through overflow. The hour-level design storm includes rainfall duration and maximum 1-h rainfall that vary with design target and design rainfall, which is an advantage over a day-level design storm.