As everyone learned in high school and promptly forgot, when a gas is at some given temperature the molecules in it have a distribution of speeds, with a long tail on the high speed end. There’s a peak speed, and the peak speed increases with temperature and the distribution widens. It is fairly easy to describe and is given the name Maxwell-Boltzmann distribution, which should tell you how long people have known about it.
The formula works for plasmas and free electrons, which are pretty common around stars. Turns out there’s a hidden assumption in the prediction of the distribution though—equilibrium.
People who study the plasmas around the sun use a class of related distributions called "kappa" distributions, which have a much bigger high energy tail. When your gas or plasma is fed with some energy source (magnetic reconnection, or what have you) and the gas doesn’t have time to come back to equilibrium, you get more high-energy particles in the mix—the high speed tail increases.
I didn’t know that, and I didn’t know that astronomers were scratching their heads about the temperature estimates; but Nicholls did, and he put the two together in one of those "obvious" connections. If you do the calculations for a "kappa" distribution instead of a "M-B" one, you find that processes that sample the high speed part will proceed as though the temperature were higher than it really is, and those sampling the lower speed part will proceed as though it were lower. So the team goes on to estimate "kappa" for several sources, and note that several have good candidate for providing a strong source of high energy electrons.
I like it. A "cool" result, and it sounds plausible.
A subscription is required to view the article itself, though unless you have some professional interest in it you probably shouldn't bother.