Elucidating the performance of hybrid models for predicting extreme water flow events through variography and wavelet analyses

Accurate prediction of extreme flow events is important for mitigating natural disasters such as flooding. We explore and refine two modelling approaches (both separately in combination) that have been demonstrated to improve the daily peak events. These are firstly, models aggregate fine resolution (sub-daily) simulated from a process-based model (PBM) daily, secondly, hybrid combine PBMs with statistical machine learning methods. propose use variography wavelet analyses evaluate these across temporal scales. exploratory methods applied both measured modelled data order assess performance latter capturing variation, at different scales, former. compare change points detected by analysis (measured modelled) identified data. found combining improves finer but coarser scales advantages less pronounced. Although aggregating fine-scale outputs improved partition variation autocorrelation signal well represented variography. demonstrate time-series analyses, using variograms wavelets, provides useful assessment existing newly proposed models, respect how they capture changes variance also this correlates – all context