Answers are grounded in this paper and cite their sources.
Nowcasting tool N‐FOREWARNS shows effectiveness in predicting surface water flooding, implying future improvements in accuracy.
Surface water flooding is generally caused when intense rainfall overwhelms drainage systems and/or floods areas before reaching a major watercourse. A key driver of surface water flooding is rapidly developing, localised convection, which is poorly represented in numerical weather prediction models. Nowcasting uses current atmospheric observations to make more accurate predictions of future convective activity on hourly timescales, making it a well‐suited approach for surface water flood (SWF) prediction. This study presents Nowcasting‐FOREWARNS (N‐FOREWARNS), a development of the daily Flood fOREcasts for Surface WAter at a RegioNal Scale (FOREWARNS) tool, which has been modified here to produce hourly SWF nowcasts for lead times of 1–6 h. User feedback from a nowcast comparison testbed indicated that N‐FOREWARNS produced few false SWF predictions. When predictions were made by N‐FOREWARNS, they were reported to have high spatial accuracy, with users noting an incorrect location only 7% of the time. However, users also reported that N‐FOREWARNS missed a high proportion of smaller‐scale SWF events with minor impacts. Quantitative verification shows that N‐FOREWARNS's 1 h lead time nowcasts produce useful skill, comparable to FOREWARNS, but N‐FOREWARNS reaches a limit of predictability at 3 h lead time, and that the skill is primarily limited by the accuracy of its rainfall inputs. Overall, these results show promise for nowcasting regional‐scale SWFing and provide a benchmark for further development, which should focus on improving the accuracy of convective rainfall inputs.
No takes yet. Share an insight, caveat, or question.
Smith et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: