A flood-crest forecast prototype for river floods using only in-stream measurements

Abstract Streamflow forecasting generally relies on coupled rainfall-runoff-routing models calibrated and executed with data estimated by monitoring protocols that do not fully capture the dynamics of unsteady flows. This limits ability to accurately forecast flood crests issue hazard warnings. Here we utilize directly measured datasets acquired for streamflow estimation develop a data-driven algorithm does require conventional physically-based modeling. We test potential our using measurements at an index-velocity gaging station Illinois River, USA, between 2014 2019. find protocol is able deliver short-term predictions crest magnitude arrival time. The produces better agreement larger events more reliable single-peak storms possibly due prominence hysteretic behavior in such events. conclude can be stage alone.