<p>Data assimilation (DA), the science of fusing different observation sources to predict possible statistics of a dynamical system, originated from the field of numerical weather prediction and later was applied for applications in geoscience, geomechanics, and engineering. In the past decade, DA has received extensive attention in experimental fluid dynamics, with typical applications spanning from data analysis and error reduction to measurement data augmentation. The predictive (physical) model used in DA is critically important, differing from the data-driven approaches in machine learning. This review provides a basic understanding of the DA methodology, the mathematics involved at the grassroots level, and the various applications in the fluid measurement community.</p>