Impact Assessment for Rainfall-Driven Inlet Clogging in Urban Inundation Modeling: A Pilot Study of an Event-Based Reverse Computing Approach

Rainfall-driven inlet clogging is a phenomenon where inlets of stormwater networks are covered with debris, fallen leaves, or trash flushed by street runoffs due to heavy rainfall. Surface runoffs are thus unable to enter sewer networks that are still under capacity, leading to greater flood risks. To provide a feasible way to reflect such inlet clogging in pluvial urban inundation modeling, a new event-based reverse computing approach using simulated and surveyed data is proposed. The performance of this approach in the aspect of accuracy improvement is then discovered in a flood-vulnerable area. The optimal way for delineating clogged inlets is first explored. Next, the possible range of clogging factors used to represent capacity-reduced effects in inlets is decided by comparing the simulated and reverse-computed flood volumes. The impact of inlet clogging on urban inundation physics is also assessed. The results indicate that, for the study site, considering inlets surrounded by trees to be 10–40% clogged (median of 25%) will increase overall accuracy. Furthermore, inlet clogging reduces drained discharges entering sewer networks, thereby increasing surface water volumes by about 10–20%, diminishing spurious sewer surcharge, and decreasing sewer water depths. Interestingly, greater drained discharges are found in clogged inlets at local depressions, even though most clogged inlets have smaller drained discharges. In summary, the present approach is demonstrated to be a useful tool to account for rainfall-driven inlet clogging in urban inundation modeling.