The alignment votrex core alogrithm as described in On the Velocity Gradient Tensor and Fluid Feature Extraction is used on the tapered cylinder data by Jespersen and Levit.

This approach uses vorticity instead of the eigenvector associated with the real eigenvalue in the swirling classification. This technique has the effect of removing the bulk and shear components of stress from the analysis. The result of this algorithm is, at times, longer and more contiguous cores. For example, tracking a horseshoe vortex in a turbine cascade, the core line would disappear as the flow accelerated around the crown of the suction surface. This was due to a change in classification from swirl to non-swirling flow. With the irrotational components removed the core can continue to be extracted. The influence of the irrotational terms also explains why the core segments would break-up when the core strength would diminish. Under these conditions the stress components end up corrupting the direction of the eigenvector used for the swirl axis.

Data courtesy of NASA Ames Research Center.