| || |
By now I'm sure that you are at least minimally aware that I have been working on the BARREL team. And if you aren't aware you are now! BARREL, if you didn't know, was a fantastic little balloon mission where we flew over 50 balloons, 40 in Antartica during the primary mission and 14 in Sweden during our no-cost extensions. Our primary science objective, the thing we were paid to go study, was to study the loss of radiation belt particles to the upper atmosphere. We were hoping that since we were launching during the solar maximum there would be a ton of geomagnetic storms and lots of activity for us to study...
Wow, I didn't realize how far behind I am with writing up blurbs on my more recent papers. I apparently haven't written anything since July and there have been three papers which have come out since then. So this will be the first in a hopefully quick series of blogs about these results. The paper we'll discuss today harkens back to my graduate school days. It's a paper on the CRRES mission: Dependence of EMIC wave parameters during quiet, geomagnetic storm, and geomagnetic storm phase times, Halford et al 2016 Journal of Geophysical Research.
When you are completing a PhD you typically end up dedicating more years than one would like to admit learning about, studying, discovering, and obsessing over a very small section of the edge of our understanding. The topic of my thesis was occurrence and plasma characteristics of Electromagnetic Ion Cyclotron (EMIC) waves on the CRRES mission during different geomagnetic conditions. Notice it wasn't EMIC waves in general, it was EMIC waves on this one satellite. It wasn't wave properties, or generation mechanisms, propagation of or a whole host of other important factors for EMIC waves, it was one small specific topic... But by the end of my PhD I was the expert on EMIC waves observed on CRRES and their occurrence during different types of magnetospheric weather. Granted this did make me one of the experts on EMIC waves in general, but that's not saying much when there are perhaps only a hand full of those in the world anyway.
So this paper goes back to much of the work that went into the thesis. It for the most part summarized all of the results into one big picture. From the first glance, this looks like and incredibly boring paper filled with lots of statistics and that's about it (there are 55 tables in the supporting information). But I think it's probably one of the more important papers that I've written. I say that because it's the start of building an empirical model of EMIC wave parameters and local plasma characteristics which are important for wave-particle dynamics. In most current global magnetospheric models, these and many other wave types are unable to be self consistently (occurs from just the equations used in the computer model) included. What does that mean... Well when we try to model the magnetosphere, the space is so large, our grid and time steps have to be relatively large. Any physics, any activity that takes place on shorter time scales, or smaller spatial scales can not be resolved. If it can't be resolved then we can't include it.
Think of taking a landscape photo. Like this one of BARREL.
You can see the BARREL balloon floating away over the lake. It's a bit small, but you can clearly see it. You can sort of make out that there is a payload and all. So lets imagine that this picture is equivalent to our picture or global models of the magnetosphere and EMIC waves are the GPS antenna. If we zoom into the photo we definitely can't make out the antenna. The pixels do not give enough spatial resolution to be able to resolve even the full box itself.
However, even though EMIC waves are small, they have a large impact... just like how without our antenna we wouldn't be able to get any of our data down from BARREL. So how can we account for the EMIC waves in our models? We try to state when, where, and over what region we expect EMIC waves to occur and then put their affects in by hand. This of course allows us to include the correct physics, even if our models can't.
Now you may ask "why can't we just make our models have smaller time and space steps?" This is often my question to modelers. They assure me, and once you look at it you can easily see yourself, that if we were to do that, the computations would take months, years, even decades in some instances instead of the days and weeks the runs take now. There are some modelers who do increase the grid and time steps in order to model EMIC and other waves. But they do this in very small regions, so their models aren't global. We keep working on ways to improve our models, but it will be quiet awhile before we can improve them to the point of having EMIC waves in there self consistently (not having to force their occurrence but let them grow and die naturally in the model itself). Really we need an advancement of computers themselves...