Isosurface of a Cross Section over Time

Now we can render the isosurface of the scalar field on one cross section over time. Several snapshots are shown below for illustration. The testing scalar field is PHI. Meanwhile, our system also allows the view to apply clipping planes to see the internal portion. By using a clipping plane that is perpendicular to the time axis, the viewer can see the cross section in different time step to understand how the scalar field varies at the time step.


PHI= +5.0e-7


PHI= +3.0e-7


PHI= +1.0e-7


PHI= +0.5e-7


PHI= -0.5e-7


PHI = -1.0e-7


PHI = -3.0e-7


PHI = -5.0e-7

Visualization of a cross section over time

Now we just create a visualization of the GTC dataset, which is to render the scalar field on one cross section over time via volume rendering. One rendering image is shown in the left top. Here the group ID is 31, the scalar field is the electrostatic potential (phi). The color is linearly mapped from the scalar to a color space, where a portion near zero is transparent. In this image, the time axis is aligned with the horizontal axis in the image, where the left(right) side means the earliest(latest) of the simualtion.



This image can be compared with its histogram of the scalar field over time, which is shown in the left bottom. Similar to the left top image, the horizontal axis represents the time, where the left(right) side means the earliest(latest) of the simualtion. Each column in this image then represents the histogram at one time step. The histograms are aligned so the entry including scalar zero is vertically aligned to red horozontal line in the image, and each histogram is draw as a vertical white bar, whose vertical length is protional to the scalar range of the corresponding time step. The time histogram clearly shows that after the first half of the simulation, the range of scalar field becomes stable in a limited range. Consequently, when assigning color/opacity to the scalar field, we can mainly consider the scalar field near the end of the simulation.

Video of both heatflux and electrostatic potential (phi)

This video contains the visualization of both iron heat flux (top row) and electrostatic potential (bottom row). The left and right columns were rendered by using linear and log color mapping, respectively.

Initial Images: Correlation between Potential and Ion heat flux

The images below are a collection of thumnails taken from 32 timesteps in the middle of the simulation. The images are ordered from left to right, and top to bottom in time. In each image, x-axis represents potential and y-axis represents ion heat flux. A periodic expansion and shrinking of regions with dense data points can be observed.

The shape of the regions with dense data points is similar to a pyramid. The following three images are generated from a cross-section of the mesh that corresponds to the dense regions shown above. From top to bottom, the images are corresponding to an early, a middle and a later time step.

Meeting Minutes (08/31/2009)

• COMMENTs about current visualization result
• The videos of both color mapping schemes (linear and log) can be placed side by side for comparison. This will provided the user with the ability to identify which color mapping is suitable for which time step. Probably linear mapping is more appropriate for a range of time steps and log mapping is suitable for the rest.
  
• The color mapping can only consider the scalar inside the limited range, eg, 95% of the scalar range, since the scalar distribution should be closed to Gaussian distribution.
• Currently, the rendering of the cross section is more effective to them than the volume rendering of the entire toroid.
  
• NEXT STEPS: the next variables/objectives to visualize:
• The particle, which is in the 6 dimension (3D for location + 2D for velocity + 1D for time)
• To visualize the trajectory of each individual particle
• To visualize its correlation of the particle motion with Eddy.This is actually an interesting visualization problem since the the eddy (4D: 3D space + 1D time) and the particle (6D) are in different dimensions. How to see the correlation between 2 variables coming from two spaces with different dimensions? Will it be more approachable if the two variables are reduced to the same dimensions?
  
• Ion Heat Flux, which is indirectly related to the particle motion since the heat flux carries the particles.
• They are interested to look at the heat flux for a fixed cross section over time. Time constitutes the third dimension of the hypothetical volume where the cross-section itself is 2D.
  
• The correlation between the heat flux and the eddy.
• Create iso-surface of heat flux over time and calculate the related quantities (e.g. the internal volume and the moving direction)
• TO DO
• Create a blog to share information, feedback, and comments
  
• Study references
  
• The measurement of quantitative information of iso-surfaces (eg. Contour spectrum)
• Multivariate analysis and visualization
  
• QUESTIONS
1. What's the index of heat flux? So far we only know that phi is the first variable.
2. Where to get the particles data? Is it included in the data we already have?
3. Which variable's isosurface and which iso-value should be looked for?
4. Which cross sections are more important?

Volume Rendering w/ Both Color Mapping Schemes Side-by-side

This video shows the cross section between the planes y=-0.002 and y=+0.002 with both color mappings (Left: linear; right:logarithm). The color red(green) represents negative(positive) scalars. The color becomes blue for color closer to 0. The domain of the color mapping is [-0.5e-6, +0.5e-6], where 0.5e-6 is half of the global maximum.

Welcome !!

Hi,
Welcome to this virtual round table which we can use to publish any information including results, meeting plan, meeting minutes related to the GTC data visualization project. Experts' feedback and thought-provoking comments that will eventually lead to brainstorming sessions are most invited.