I am now finished with my research endeavor at Fermilab and have returned to campus for the semester. I certainly learned a lot simply by osmosis from being surrounded by my collaborators. Working remotely has been constructive, but moves much slower when one doesn’t have immediate access to helpful experts.
The larger aim of my project, to measure pion cross sections, is not complete, but this was not the goal of this trip. During this trip I aimed to develop my intuition for the sophisticated simulations that I must use in order to evaluate the systematic uncertainties of my research. In this I found success. I simulated photon events and learned more about the particle physics processes of photons in our detector and was able to develop a technique to bolster the efficacy of our reconstruction with the help of my collaborators. The technique involves using the underlying “tracks” (relatively straight portions likely from a single particle) as the backbone of a shower. My algorithm takes the bones of two showers and tries to figure out if they are aligned in any way, indicating that the two showers may have actually come from the same photon (it is often, about 40% of the time, that the standard algorithm returns more showers than the number of photons. This method is designed to merge showers together to ensure that there is a 1:1 relationship between the number of showers and the number of photons.) I use tracks because their straight behavior means there is a defined direction to their motion (in comparison to showers, which contain many tracks and are therefore difficult to assign a direction to). If a shower has multiple tracks moving in different directions, which way is the shower moving? Or, in the bones analogy, it might be difficult to known wether a person threw a baseball if you only know the vector of their center of mass, but if you know how their arm was moving this might inform you of if they were pitching a baseball. Moving forward I plan to simulate a larger (perhaps several hundred thousand?) dataset of photons. This will allow me to develop a probabilistic relation between two important quantities: the true momentum of a photon and the reconstructed momentum. If I can develop this probabilistic relation using simulations I will then be able to apply it to real data and answer an important question: if a given amount of energy was seen by the detector, what is the most likely true energy of the particle? If I can reliably reconstruct a true energy then I can confirm if a particle with a certain mass decayed (recall E=mc^2), and that mass will tell me the identity of the decaying particle. Thus, I return to William and Mary with a more developed sense of the tools of particle physics that will allow me to make a physics measurement important to our experiment.
Finally, a few extra things from this trip. I was also able to attend a seminar on careers in physics while at Fermilab, another way that being on site presented numerous and even unexpected opportunities. I found this presentation very interesting because previously I had no idea what I would do with a physics Ph.D. other than teach. Several Ph.D.s and former national laboratory employees left academia for the insurance company Allstate and now help make up a data analytics team. I never would have guessed that there was a need for particle physicists at an insurance company, but they have a need for people capable of handling large datasets. I was additionally able to attend our weekly meeting in person on Friday and meet more of our collaborators and see the direction of our project in a larger sense. We were planning on me helping out with an installation of some new equipment, but that had to wait on another group and their job was not complete while I was at Fermilab. Maybe next time!