27 February 2017

First step - setting the research question

So, the first step in scientific inquiry is figuring out what I wish to research. In this particular case, I have it easy, because I’m continuing my previous research, I already have a question.
I noticed there is an oscillatory response on the flare in distant regions. So I’m continuing to dig deeper.
Let me first explain what I mean by the previous paragraph. Oscillatory response - means I noticed parts of the Sun intensify their regular shakes very quickly after the flare, even when that flare is on the other side of the Sun.
The problem with this is that there is no currently good enough explanation of what transferred the disturbance over such big distance. The explanation that stem from previous knowledge, the seismic oscillations and the transfer of the particles through flux tubes are slower than observed response. So, it has to be something else.
I have an idea, and the first step is to discuss that idea with another scientist. Because I need to find the holes in my logic. Something that is very hard to notice by oneself. Luckily, a good colleague of mine is also interested in this question so I will be able to brainstorm with him.
In the meantime, I will enlarge my initial sample including all flares happened on the day. So, the next step is to get the data.
The data are easy to obtain, every publicly funded instrument is obliged to share data, one can simply demand it. There are nicely developed methods to get data.
I decided to start with Solar Dynamic Observatory satellite data, using their instrument AIA, and spectral line 171. That is what I used for the first scientific paper, so it makes sense that this starts from the same point.
And now, I’m getting to problems. Each research, each project has them. Nothing goes smoothly.
See, a majority of coding and analysis in Sun research is done using IDL code language. If you never heard of it, do not worry. This is proprietary software, made exclusively for image processing and used by Solar and medical communities. Nowhere else. Because of such a narrow audience, it is expensive (it costs about $2000 to get a license for it). And because of a captive audience, it is not very good either. Last time I used it, I could not do any parallel computing with it. Instead, I had to write codes in C++ and Phyton.
Since I’m not paid for this project, I have to skip IDL and use Phyton instead. Luckily, there are people in the Solar research community who decided to switch to Phyton too. They developed a module called Sunpy. And my first task is to read up on how to use that particular module and pre-process my data for analysis.

No comments:

Post a Comment