A tracer-aided criterion to discretize pluvial and fluvial flood hazard maps in catchment scale shallow water models

Knowing the nature of the flood hazard is a crucial factor for improving the resilience of the urban areas, since awareness, preparedness and early warning systems are based on the scientific tools such as the 2D depth-averaged shallow water models. Inland flood hazard primarily stems from pluvial and fluvial inundations, typically modeled separately respecting the pertaining spatial domains of the assessment, namely the urban areas and the riverine floodplains. Considering the high computational power and efficiency of both hardware and codes, the catchment scale hydrological-hydrodynamic modeling is becoming an increasingly adopted approach in flood hazard assessments. Since a complete rainfall-induced routing is preserved, these simulators determine fluvial, pluvial and compound inundations caused by heavy storm events within the entire watershed. However, this approach leads to flood extent maps in which the inundations such as those resulting from pluvial and fluvial processes, are usually not differentiated, even if significant disparity in the space-time scales and volumes of water are involved. Indeed, these two hazards follow distinct normative and regulatory flood risk management rules among different countries.  With such a rationale we established a tracer-aided criterion to systematically categorize and map pluvial and fluvial hazard in a catchment scale shallow water model, exploiting the advection process of a conservative tracer. The physically based methodology, implemented in the GPU-parallelized Iber+ software and its water-quality module (IberWQ+), is applied in a small urban catchment for multiple probabilistic scenarios. The results demonstrate the effectiveness of nesting transport and shallow water equations, univocally discretizing the two inundation sources in function of the computational cells reached by the tracer. This enables to define the spatial domains of the pluvial and fluvial processes, providing valuable insights for holistic catchment-scale flood risk management. Additionally, the advancements achieved by the proposed method are showcased in comparison to commonly employed modeling techniques for mapping fluvial inundations. As the tracers continue to improve our understanding of catchment sciences, we conceptualized them role through an abstraction that can aid surface hydrodynamic modelling to identify pluvial and fluvial sources of hazard.