# Stellar Evolution in Real Time: Models Consistent with Direct Observation of Thermal Pulse in T Ursae Minoris

## TL;DR

This paper combines advanced stellar modeling with long-term observational data to directly observe and analyze the thermal pulse in T Ursae Minoris, providing unprecedented precision in its fundamental parameters.

## Contribution

It presents the first direct observation of a thermal pulse in a star using combined modeling and observational data, achieving the most precise mass and age estimates for an AGB star.

## Key findings

- T UMi's rapid pulsation change is caused by a recent thermal pulse.
- T UMi transitioned into a double-mode pulsation state.
- Predicted pulsation period will shorten then lengthen over decades.

## Abstract

Most aspects of stellar evolution proceed far too slowly to be directly observable in a single star on human timescales. The thermally pulsing asymptotic giant branch is one exception. The combination of state-of-the-art modelling techniques with data assimilated from observations collected by amateur astronomers over many decades provide, for the first time, the opportunity to identify a star occupying precisely this evolutionary stage. In this study, we show that the rapid pulsation period change and associated reduction in radius in the bright, northern variable star T Ursae Minoris are caused by the recent onset of a thermal pulse. We demonstrate that T UMi transitioned into a double-mode pulsation state, and we exploit its asteroseismic features to constrain its fundamental stellar parameters. We use evolutionary models from MESA and linear pulsation models from GYRE to track simultaneously the structural and oscillatory evolution of models with varying mass. We apply a sophisticated iterative sampling scheme to achieve time resolution $\le10$ years at the onset of the relevant thermal pulses. We report initial mass of $2.0\pm0.15\, \mathrm{M}_\odot$ and an age of $1.17 \pm 0.21$ Gyr for T UMi. This is the most precise mass and age determination for a single asymptotic giant branch star ever obtained. The ultimate test of our models will be the continued observation of its evolution in real time: we predict that the pulsation periods in T UMi will continue shortening for a few decades before they rebound and begin to lengthen again, as the star expands in radius.