ATLAS jet trigger performance in Run 2 and searching for new physics with trigger-level jets

Hadronic jets play a vital role in the search for physics beyond the Standard Model at particle colliders. Jets are the most commonly produced collision products at hadron colliders like the LHC, and thus represent an extremely promising avenue for discovery. While this prevalence allows for Standard Model processes to be measured to extreme precision and exciting new physics to be searched for, analysis using jets comes with many challenges. The high rate of collision events that produce jets naturally leads to a large background that must be sorted through so that the most interesting events may be identified and recorded. This requires robust jet reconstruction, calibration, and triggering in order to maximize the amount of interesting jet events that can be analyzed for potential new discoveries. In this thesis, an analysis of the performance of the ATLAS jet trigger system during CERN LHC Run 2 is presented. Jet Trigger efficiency curves are analyzed for the full suite of HLT triggers from each year of Run 2 data collection, with differences in run conditions from year to year reviewed and their impacts on jet trigger efficiency analyzed. The effects of an improved jet calibration scheme implemented midway through Run 2 on jet trigger efficiencies and the fraction of recorded fully efficient data that is most useful for further physics analysis are investigated. Additionally, a search for new physics using dijet events at the trigger level is presented. Results of a partial Run 2 analysis using data collected through 2016 are presented and contributions to an upcoming full Run 2 analysis are discussed.