In this talk, we will discuss how data assimilation methodologies may enable combining experimental, high-fidelity and low-fidelity numerical simulations, and thus achieve complementarity amongst these approaches to study and characterize aerodynamic flows. From the experimentalist’s point of view, it will be illustrated how data assimilation may be used to increase the spatio/temporal resolution of experimental data and infer non-measured quantities, considering in particular the case of optical techniques such as Particle Tracking Velocimetry. From a more modeling-oriented perspective, we will consider the application of data assimilation to enhance Reynolds-Averaged Navier-Stokes (RANS) models based on velocity or parietal pressure measurements. Based on the so-calibrated models, stability and resolvent analysis may then be performed to predict detailed flow features that are difficult to characterize from experiments or simulations alone. Such an approach will be applied in particular to the characterization of stall cells for the flow around a NACA4412 profile at a chord-based Reynolds number of Re=3.5x10^5.