I love thunderstorms. I love the smell in the air, the sound of rain on the roof, the clean feeling everywhere when the storm has passed. I also love lightning and thunder, which never fail to move me. I'm not alone; in the ancient mysteries a device called a ceraunoscope was used to simulate lighting and thunder in religious ceremonies.

I decided to try to simulate these correctly on the computer. Building on the work of some atmospheric scientists, I've developed a statistical model of lightning that matches the natural shapes, which can be controlled by a designer to strike any location. From this 3D lightning data, and the 3D location of a listener, I compute the thunder that listener would hear due to that specific lightning bolt, including such effects as atmospheric refraction, wind, and multiple lightning flashes.

On the left we see the basic strategy for making realistic lighting. I create a "skeleton" for the basic stroke that uses real atmospheric data to compute the lengths of each piece, the frequency of branching, and the angle of branching. These all vary along the stroke as a function of height from the ground.

On the right are three different strokes produced by this method. They are, statistically, indistinguishable from a natural stroke. They have the added benefits of existing in a computer modeling system, and can be controlled and placed by a designer or filmmaker. In a sense, these lightning bolts are actors or props just like any other controllable part of a scene.

Thunder is created as a result of the intense heating of the atmosphere by a lighting stroke. I use the 3D geometry of the stroke, and the properties of the layers of air, to compute the sound that an observer would heard from a specific lightning bolt from a specific place on the ground. This figure shows the pressure signature, or sound wave, created by the lightning stroke above.

You can find full details and simulations data in the papers referenced below. It may be interested to note that I've been able to turn the process around, and create lightning as a result of thunder.

This is a shot of my lightning designer. In the upper-left I've specified a particular kind of thunder that I want to hear, by identifying the type of sounds over time. I use the standard terms for the sounds of lightning that are used by researchers in the field.

This particular graph says that I want a bit of rumble (the low green box), followed by a sustained thunderclap (the higher red box) and then later, some roll (the yellow box). The sound wave below the design graph shows one of the many pressure signatures that create this kind of thunder. Notice the tall spike at the beginning of the thunderclap. Finally on the right we see a 3D lightning stroke that has been created to these specifications.

If you use this lightning stroke in a 3D scene, and compute the actual thunder it would produce, the sound will look something like the lower-left image, and will have the right amounts of rumble, clap, and roll, in the right order and at the right times.

Glassner, Andrew S., "The Digital Ceraunoscope: Synthetic Thunder and Lightning", Microsoft Resarch Technical Report MSR-TR-99-17, April, 1999.

You can download this report directly from the Microsoft Research webs site by typing in this URL, or simply clicking on it:
http://research.microsoft.com/scripts/pubs/view.asp?TR_ID=MSR-TR-99-17

The IEEE citations are:

Glassner, Andrew S., "The Digital Ceraunoscope: Synthetic Lightning and Thunder, Part 1", IEEE Computer Graphics & Applications, Volume 20, Number 2, March 2000

Glassner, Andrew S., "The Digital Ceraunoscope: Synthetic Lightning and Thunder, Part 2", IEEE Computer Graphics & Applications, Volume 20, Number 3, May 2000