##########################
VI. Example uses of BASE-9
##########################

In Section IV we outlined how to use the outputs of BASE-9. Here we provide additional examples from our papers and on-going work.

The first figure of this section, taken from DeGennaro et al. (2009), shows Hyades CMDs with three sets of stellar evolution models placed at their average fit values as determined by a previous version of the code, BASE-8. Because these stellar models do not provide good fits to the lower main sequence, the following figure shows the derived age from BASE-8 for each of the three input models and for a range of lower main sequence cut-offs. In this way DeGennaro et al. were able to argue that their derived parameters were stable over an appropriate range of data and were able to quantitatively point to where problems emerged in the models.

.. image:: charts_2.png
    :width: 735px
    :align: center
    :height: 511px

.. image:: charts_3.png
    :width: 570px
    :align: center
    :height: 284px

This next figure, taken from Jeffery, E. J. (2009, Ph.D. Dissertation, University of Texas at Austin) compare the age information resident in just the main sequence turn-off stars (black dashed line) compared to that resident in the white dwarfs. Data from the main sequence was included in both BASE-8 runs, and this provides the primary constraints on metallicity, distance, and reddening. This is useful for studying the information content in the MSTO vs. WD regions.

.. image:: charts_4.png
    :width: 611px
    :align: center
    :height: 586px

The next figure, also from Jeffery (2009), indicates how one can study the sensitivity of a given result to the observations of an individual star. For the open cluster NGC 2360, the posterior age distribution is given by the black line. During some iterations, however, a particular WD is fit as a field star and the remaining WDs yield the posterior age distribution indicated in red. During the iterations when this particular WD is included in the fit, the posterior age distribution is as indicated in blue. The final age posterior distribution is a linear combination of these two distributions based on the fraction of time this particular WD was included in the fit.

.. image:: charts_5.png
    :width: 990px
    :align: center
    :height: 966px

The next figure shows unpublished work based on applying BASE-8 to an individual WD. In this particular case, we know that the WD has a hydrogen atmosphere (type DA), yet for demonstration purposes we analyze it both as a DA and as a DB (helium atmosphere). We also try two different initial-final mass relations (from Salaris et al. 2009 and Williams, Bolte, & Koester 2009). The clouds of points show acceptable fits and the error bars indicate the mean and standard deviation for each of the four cases. Clearly these distributions are non-Gaussian and publishing just the means and standard deviations could lead readers to misunderstand the results. This kind of analysis can also point the way toward future observational work. For this star a trigonometric parallax could potentially rule out much of the age range, yielding a precise age. If this star were a DB a much more accurate trig parallax would be required to meaningfully constrain the age. This is not a general statement about WDs, but a result for this star with the available photometry (grizJHK).

.. image:: charts_6.png
    :width: 639px
    :align: center
    :height: 540px