In this paper, we propose the application of standard decomposition approaches to find local correlations in multimodal data. In a test scenario, we apply these methods to correlate the local shape of turbine blades with their associated aerodynamic flow fields. We compare several decomposition algorithms, i.e., k-Means, Principal Component Analysis, Non-negative Matrix Factorization and Uni-orthogonal Non-negative Matrix Factorization, with regards to their efficiency at finding local, relevant correlations and their ability to predict one modality from another.