12 June 2017

Hmm, that's funny...

I know that you are all taught that Eureka is saying which leads to significant discoveries. But the truth is, it is more likely for a scientist to say 'That’s funny' than 'Eureka.'
During the data analysis we keep our eyes open to figure out what is going on, and when something out of the expected appears, we do whatever we can to explain it.
My first run of the data analysis is almost done. As I explained previously, I needed to establish a baseline for the detected oscillations so that I can see what is going on when there is a flare.
And during this initial step, I uncovered an interesting effect. See, this is how the power map should look like when everything goes as expected.
Screen Shot 2017-06-11 at 16.02.59
But for some areas, I got unusual noise.
Screen Shot 2017-06-11 at 16.08.32
And yeah, this caused me to go, "Hmm, that's funny." Unfortunately, this funny result does not lead to some breakthrough discovery. In fact, the shakes come because in the image I have magnetic flux tubes that move across the field of view. This happens because some of the field of view I chose is hanging on the left or right side of the visible solar disk. And since the Sun is roundish you get a different angle of view when you look at something located in the center than something located on the sides of the visible disk.
Flux tubes appear like tufts of hair on the Sun's surface. And a hair looks different when you look at it from the tip down than when you look at it from the side. No brainer really.
The movement gives an impression of more power happening in the image, but in a way, this is just movement. So this example neatly illustrates the importance of knowing exactly what your tool does. Mine translates the slight variations in the light of the pixel to the oscillations. This is perfect if you observe something that faces directly at you and has an oscillation that heads straight towards you. Any angle messes up the result.
And some of my areas have an angle. So why to bother with the tool that does not give correct results, you might ask. The answer is because I’m impatient. I also know that the same tool will discover the movement, and I would like to see the movement of the flux tubes. So for me, these strange noise-like shakes are actually good news. I am happy that this relatively straightforward and fast analysis can discover the movement. Of course, for the scientific paper, I will have to make sure to get correct measurements of the movement, and that means making a slightly different code, i.e. using a different tool.

No comments:

Post a Comment