Visualization

Examples of visualizing ACCIV output

The two final output files from ACCIV are outScatteredVelocity.h5 and outGridVelocity.h5.

plotVelocities.py

There are a variety of ways of visualizing the resulting data. The Python file tests/syntheticTest/plotVelocities.py creates 10 different plots based on the results of the syntheticTest (obtained by running either sixPassScript.csh or sixPassScript.sh). These plots can easily be adapted to other data sets.

The plots include:

1. a quiver plot of 5000 scattered velocity vectors chosen randomly form the data set
2. a quiver plot of every 8th velocity vector in each dimension from the gridded data set
3. a 2D map plots of vx from the gridded velocity data set
4. a 2D map plots of vy from the gridded velocity data set
5. a 2D map plots of |v| from the gridded velocity data set
6. Histograms of vx, vy and |v| (useful for verifying the searchRange parameter, see Setting Up Input)
7. a slices of the scattered vy data through the center of the image along the x axis
8. a slices of the scattered vx data through the center of the image along the y axis
9. a histograms of the correlation velocity uncertainty (see Calculating Uncertainties)
10. a histograms of the correlation location uncertainty (see Calculating Uncertainties)

Further examples

Here are some examples of visualizations of ACCIV output, taken from various data sets:

This is velocity map. East-west velocity is shown. For maps, it is best to use data in outGridVelocity.h5. Maps of north-south velocity can be displayed the same way.

This is a scatter plot. Sometimes we call this a "quiver plot" because an archer’s basket of arrows is called a quiver. This is a good way to plot the data in outScatteredVelocity.h5 (Asay-Davis et al. 2009, Fig. 2b).

The `tests/syntheticTest/' project contains both Matlab and Python scripts for performing visualization of both scattered and gridded velocity data.

These are vortex principal axis plots. These could also be called north-south and east-west velocity cuts, or “spoke plots” (particularly if there are other angles besides 0° and 90° from the east-west axis). These types of plots should be constructed with the data in outScatteredVelocity.h5 (Asay-Davis et al. 2009, Fig. 4).

The three plots above are the most basic types of visualizations, without which it is difficult to evaluate ACCIV output and figure out how to modify parameter settings.

Histograms can also be useful to look at:

This is a histogram of uncertainties in the velocity field, using the correlation velocity uncertainty metric.

This is a histogram of velocities in the outer high-speed ring of Oval BA in 2000, 2006, and 2009 (Wong et al. 2011. http://dx.doi.org/10.1016/j.icarus.2011.06.032)