<p>Accurate estimation of design floods that capture intra-seasonal variability in magnitude (peak, volume, duration) and hydrograph shape (prior/posterior-peaked or multimodal) is critical for enabling dynamic and adaptive reservoir flood-mitigation operations throughout the flood season. In this regard, the current study adopts a statistical approach in an integrated framework to derive intra- seasonal design flood hydrographs for pre-defined flood sub-seasons such that the severity of the floods in terms of their magnitude as well as shape is captured simultaneously. This study applies Vine copulas to model complex multivariate dependence among key flood variables for design flood estimation. The randomness in hydrograph morphology is captured through normalized representative hydrographs extracted from historical records, ensuring that the inherent variability of real-world flood behaviour is retained in the analysis. For a given joint return period, the best-fit four-dimensional canonical/ drawable vine copula, identified from twelve possible vine structure configurations, is applied to generate the target design flood characteristics. These characteristics are subsequently used to upscale the typical flood hydrographs. Notably, the use of more than three variates in hydrological studies has seen only limited application, underscoring the novelty of the present approach. The choice of targeted design variables relates the flood magnitude and hydrograph shape thereby overcoming the limitation regarding the arbitrary selection of flood hydrograph shapes. The proposed framework serves as a robust tool for seasonal design flood estimation, demonstrated using flow records from the upstream reach of the Nacimiento Dam, California. By generating design hydrographs that more accurately reflect real-world flood behavior, the framework is expected to support the development of more effective and pragmatic flood mitigation strategies.</p>