Baroclinic Instability

The first series of experiments always work better.

When I carried out the first series of experiments, I didn't have any proper laboratory equipment, so with Ryuji Kimura's "Science of Currents" in one hand and a motley collection of materials, I somehow managed to get some beautiful clear photos of each experiment.

Encouraged by the results, I started work on some full-blown laboratory equipment (apparatus 2 and apparatus 3), but no matter how much time or money I spent on the apparatus, I couldn't get the images I wanted, and it ended up taking over ten years before I finally managed to outdo my original efforts.

Good experiments and good education don't go together?

At first we carried out the measurement of the temperature of the inner and outer sides of the water tank with a thermometer. However, there were four poles supporting the camera on the rotating table, and we had to take measurements between revolutions of the four poles. As this was a little dangerous, we made a small digital thermometer, and filmed it with the video camera.

Thinking I'd solved my problems, I was then faced with "protests against rationalization" by students who had mastered the craftsman-like skill of reading the normal thermometer between the moving poles.

Geophysics means endurance?

When obtaining the three-dimensional position of the path of the nylon ball recorded on video tape, you have to read off the coordinates of the ball. It would be a pretty neat high-tech trick if we could use some kind of computerized image analysis, but when the experiments were carried out in 1985, that kind of technology was way out of reach. Instead, I drew the graduations of the coordinate axes on a large video monitor, and read off the coordinates with the naked eye. However, video recorders had only just started to spread to the average home, and when you paused the tape, the image wavered too much to be read. In the end, because we had to read off the coordinates at normal playing speed, we had a "reader" and "recorder" for each coordinate, and another person in charge of the stopwatch, for a total team of 7. At the signal of the timer, 3 of us read off three coordinates at once, and read them aloud in a pre-determined order to our respective recorders. By training ourselves in this method, we managed to get the sampling cycle down to two seconds, a real victory of mind over matter.

 

More Light!

When I first started experiments using liquid crystal capsules, I was using a slide projector as the source of the necessary slit of light, but couldn't manage to get a good clear photo of the changing color of the liquid crystal capsules. If there only a few capsules, the color was weak, but if I put too many in, this time the working fluid would cloud over, making the contrast even worse.

The reason for the clouding is perhaps because the body of the capsules was made of resin and had a different refractive index to the water. If so, then if we could match the refractive index of the water to the capsules, then we should be able to get a clear shot. While trying this and that to solve the problem, I came across a beautifully clear shot of the liquid crystal capsules being used to make the current visible. Without introduction or previous contact, I rang the author of the book in which I found the photo, (Prof. Hideo Kimoto of Osaka University) and received a reply that "it is simple to get a clear shot". I found that the only real difference between my apparatus and that of Prof. Kimoto was the light source I was using. I contacted a light maker and still in a little doubt as to whether it would work, I tried a different light source. It worked beautifully.

Student: "What was the reason for all that struggle when the solution was so simple?"
Me: "Well, ah... that's the way experiments are. You should be grateful for the opportunity to find these things out."