Misleading temperature variance output

[abrownless1]

Qdyn6 reports temperature averages together with their variances. For some small temperature groups, instantaneous temperature estimates fluctuate strongly, which causes the reported variances to become very large. While this behavior is statistically expected, the resulting numbers are unintuitive and can be misleading to users, as they may appear to indicate an enormous physical instability or thermostat failure.

Reporting the standard deviation instead of the variance would provide a more interpretable measure of temperature fluctuations. Alternatively, omitting temperature fluctuation statistics altogether (and reporting only averages like what was done in Q5) would also avoid confusion.