The past few years have witnessed the rise of neural networks (NNs) applications for hydrological time series modeling. By virtue of their capabilities, NN models can achieve unprecedented levels of performance when learning how to solve increasingly complex rainfall-runoff processes via data, making them pivotal for the development of computational hydrologic tasks such as flood predictions. The NN models should, to be considered practical, provide a probabilistic understanding of the model mechanisms and predictions and hints on what could perturb the model. In this paper, we developed two NN models, i.e., Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS) and Network-Based Expansion Analysis for Interpretable Time Series Forecasting (N-BEATS) with a probabilistic multi-quantile objective and benchmarked them with long short-term memory (LSTM) for flood prediction across two headwater streams in Georgia and North Carolina, USA. To generate a probabilistic prediction, a Multi-Quantile Loss was used to assess the 95th percentile prediction uncertainty (95 PPU) of multiple flooding events. Extensive experiments demonstrated the advantages of hierarchical interpolation and interpretable architecture, where both N-HiTS and N-BEATS provided an average accuracy improvement of ∼ 5 % over the LSTM benchmarking model. On a variety of flooding events, both N-HiTS and N-BEATS demonstrated significant performance improvements over the LSTM benchmark and showcased their probabilistic predictions by specifying a likelihood objective.

In recent years, numerous flood events have caused loss of life, widespread disruption, and damage across the globe. These devastating impacts highlight the importance of a better understanding of flood generating processes, their impacts, and their variability under climate and landscape changes. Here, we argue that the ability to better model flooding is underpinned by the grand challenge of understanding flood generation mechanisms and potential impacts. To address this challenge, the World Meteorological Organization-Global Energy and Water Exchanges (GEWEX) Hydrometeorology Panel (GHP) aims to establish a Global Flood Crosscutting project to propagate flood modeling and research knowledge across regions and to synthesize results at the global scale. This paper outlines a framework for understanding the dynamics and impacts of runoff generation processes and a rationale for the role of a Global Flood Crosscutting project to address these challenges. Within this Global Flood Crosscutting project, we will establish a common terminology and methods to enable the global research community to exchange knowledge and experiences, and to design experiments toward developing actionable recommendations for more effective flood management practices and policies for improved resilience. This harmonization of rich perspectives across disciplines will foster the co-production of knowledge primed to advance flood research, particularly in the current period of heightened climate variability and rapid change. It will create a new transdisciplinary paradigm for flood science, wherein different dimensions of mechanistic understanding and processes are rigorously considered alongside socioeconomic impacts, early warning communications, and longer-term adaptation to alleviate flood risks in society.