================================================================================

               "Multiscalar Mixing in Isotropic Turbulence"
                   Bruce A. Perry* and Michael E. Mueller
	                  Princeton University
            Department of Mechanical and Aerospace Engineering
 
                         *[email protected]

================================================================================

INCLUDED FILES:

perrymueller.mp4  : VIDEO ENTRY to the visualization contest
pm-still.pdf      : Still image for printing purposes 

allpdfs.py        : Python script to construct frames of the visualization 
pdfpaper.py       : helper script for above
vid.py            : helper script for above

data (directory)  : contains data necessary to reconstruct frams in the 
                    visualization
video (directory) : directory to which allpdfs.py outputs frames. containes one
                    sample frame

================================================================================

(RE)CONSTRUCTING THE VISUALIZATION :

The video contains two parts: a depiction of the cubical domain with an animation
of the velocity field, and four brief animations of three components (red, 
yellow, and blue) mixing together, with models for the mixing distributions
compared to the actual distribution of mixtures.

The first part was generated by exporting a video from the 3D data visualization
software Ensight-3D. This was done using a GUI so no code is provided to
reproduce this part. 

The animations of mixing are the main focus. Each frame is created by the python
script that extracts and plots data that was saved from a CFD simulation. A 
nonlinear equation solve is required to get predictions from some of the models, 
and this is run at the time the figures are generated. Because the datafiles 
required to generate each frame are very large, data has only been supplied to 
recreate one frame.

To recreate the frame, simply run the Python script "allpdfs.py". The frame will
be saved in the video directory. The script was written and tested using 
Python 2.7. It is depndent on NUMPY, MATPLOTLIB, and SCIPY. 
 
Groups of frames were concatenated into animations using ImageMagick, and the 
animations were assembled into the final video using Microsoft PowerPoint.

================================================================================

CAPTION:

Multicomponent turbulent mixing plays a key role in determining the behavior of
many engineered and geophysical systems. This visualization shows how three 
components (red, blue, and yellow) mix together in a turbulent flow field. The 
mixing state is represented using a triangular, two-dimensional color map.
The triangular contour plots show the distribution of states in the system as it
evolves, comparing results from the simulation to several models. It is shown 
that the initial distribution of the three components affects which models 
perform the best. This work verifies theoretical predictions that were made 
based on the mathematical properties of the model distributions.