2022MNRAS.516.4497S

U Sco is a recurrent nova with 11 observed eruptions, most recently in 2010.1 and 2022.4. I report on my program (running since 1989) of measuring eclipse times and brightnesses of U Sco in quiescence, from 2010 to 2022. The orbital period suddenly increased by +22.4 ± 1.0 parts-per-million across the 2010.1 eruption. This period change is greater than the near-zero period change (+3.9 ± 6.1 parts-per-million) across the 1999.2 eruption. This period change cannot come from any of the usual mechanisms, whereas the one remaining possibility is that the period changes are dominated by the little-known mechanism of the nova ejecting asymmetric shells. From 2010.1 to 2016.78, the O - C curve showed a steady period change that was large, with $\dot{P}$ = (-21.0 ± 3.2) × 10^–9. This is greatly higher than the steady period changes in the two previous inter-eruption intervals (-3.2 ± 1.9 and -1.1 ± 1.1 × 10^–9). This large, variable, and negative $\dot{P}$ apparently comes from magnetic braking of the companion star's rotation. Starting in 2016.9 ± 0.6, the O - C curve showed a strong kink that is a unique characteristic of the sudden period change (+35.4 ± 7.1 parts-per-million) across a nova event. The brightness in quiescence after 2010.4 shows that the white dwarf accreted the trigger mass for the next nova event in the year 2017.1 ± 0.6. Photometric records show the only possible time for the eruption to peak (such that its total duration of 60 d was undetectable by any observation) is during a 75-d interval inside the 2016 solar gap, thus constraining the missed eruption to 2016.78 ± 0.10.