FEPS PROGRAM STELLAR AGES
 Original: 5 April 2005
Last Revised: 2 August 2005 (active editing 12 Sept 2007)
Lynne Hillenbrand


For discussion by the "core" FEPS ages group:  Hillenbrand, Mamajek, Stauffer, Soderblom, Meyer, Carpenter
(as well as anyone else who can substantially participate --  make yourselves known!!)

Skip to:

......Goals ......
......Procedure to realize an age ...... 
......Project Status ......
......Database issues ......
......Age estimators ......

......Assigning final ages......


Goals:


Status:

after a two-year hiatus, the ages group is back in action!

link to current paper figures [not updated since 2005]

link to ages derived from various techniques [not updated since 2005]
figure illustrating age vs age correlations. ACTION: assess offsets and correlations with color; correct age calibrations

link to ages for "IRAC paper" targets
link to ages for "first-look" gas targets


Procedure to Realize an Age:
 plot of dispersions in R'HK, vsini, vrad


Database issues:

Status:  Much age-relevant info in FEPS database
        -- allende-prieto et al 1999
        -- nidevar etal 2002
        -- wright et al 2004
        -- nordstrom etal 2004 (spot checking reveals some data may not be entered?)
        -- valenti and fischer 2005  
        -- white et al. 2007
       
        -- soderblom R'HK values
            

Data actions:  


Estimators of stellar age:
membership in an open cluster
Relevant # of FEPS stars:  ~130
                                             22 - Hyads,  20 - Pleaids,  13-  Alpha Per,  5 - IC 2602,  
                                             14 - LCC,  23 - UCL,  17 USco,  7 Cor Aust

Status:

Data actions:


Analysis actions:

General Concerns:

hrd isochrones

Relevant # of FEPS stars: ~90 Pre MS and  ~240  MS

Status:
                                 log(age/yr)   age(Myr) 
                      LCC                           17
                      IC 2391   ~7.71          52       
                      Pleiades  ~8.01         102       
Data actions:  
           

Analysis actions:
           JRS to do small study of using observables to define CMD locations vs. using LogL, LogT from JMC database. JMC to try fixing Av for Pleiades, Alpha Per and ? in order to see how big an effect this is on L and T.   DONE  see plots

           I have a program that makes monte carlo realizations of HRD positions,    
           interpolates their ages and masses from tracks, and then quantifies the
           mean, median, and confidence intervals for the ages and masses. The
          outstanding problems are (in no particular order):
(1) the degenerate overlap between PMS and MS tracks,  
      Stauffer:  I think it is reasonable to believe that
      with few or no exceptions, other age indicators should allow
      us to break that degeneracy
(2) we need logT and logL values with errors that people agree on,
      Stauffer: my preference would
    be to plot the observables and compare them to isochrones that
    either are natively in colors and mags or that we convert to
    colors and mags.   At least the Berkeley model is to go that
    route - plot the objects we observe in the sky in the units they
    are observed as much as possible, and where necessary transform
    models to that plane.
(3) which tracks do we use?,
(4) should we "calibrate" the tracks?,
     Stauffer:  my vote is yes, given that even
    the track makers generally agree that they do not yet model
    optical colors correctly
(5) how do we report the output (median? mean w/ 68% errorbars?),
(6) do we also Monte Carlo the dispersion in [Fe/H]?,
(7) should we use prior information? (i.e. a star may lie on a 30 Gyr-old
isochrone, but it is unphysical),
(8) can John report a f_bol w/ error from the fit between the photometry
and the Kurucz models? It would seem that this may be better than using 
individual colors and magnitudes which may have systematic effects (see
Stauffer+ 2003),
(9) I need to provide "final distances" to some young field stars (I need
to enter the results of my thesis into the database), and provide
distances to stars that appear to members of groups.
    UPDATE:
* I have a file that is ~90% done with the weighted mean *trig* parallaxes
from  Hipparcos, Tycho-1, and Yale. this will probably be the best-of-the-best
 distances for ~tens of % of our sample. they won't change much from whats
in the database (usually Hipparcos), but the errors will shrink a bit, and it
incorporates two other independent parallax estimates in many cases
(tycho-1 and Yale ground-based average).  link to current version



General Concerns:


rotation (vsini and period)

Relevant # of FEPS stars: ~300 vsini's, though 40 are upper limits.  ~80 stars with periods.

Status:
 Data actions:
Analysis actions:
General Concerns:

                   


xray luminosity

Relevant # of FEPS stars: 271 from a large and relatively homogeneous database: ROSAT


Status:
Data actions:
Analysis actions:

 ACTION: ASSIGN THESE TASKS           

General Concerns:

                             
CaII HK       
               
Relevant # of FEPS stars:  147 (160 measurements) from Soderblom00 (no updates are planned).  
                                           103 (111 measurements) from Wright et al. 2004.
                                            278 from White et al. 2007.
                                             

Status:
 Data actions:
 Analysis actions:
 General Concerns:

 LiI 6708
Relevant # of FEPS stars:  ~315!

Status:
 Data actions:

 Analysis actions:
 General Concerns: 

                              
HI alpha
Relevant # of FEPS stars:  ~230
                                            main reference is P60 data since we have not tabulated this quantity from the literature.

Status:
 Data actions:

Analysis actions:


General Concerns:
                          
uv continuum/line
Relevant # of FEPS stars: none at present

Status:
 Data actions:
Analysis actions:

General Concerns:

                             

               
kinematics
Relevant # of FEPS stars: ~100

Status:
                                  - 1) ALL FEPS SOURCES SHOULD BE DEMONSTRLY "YOUNG DISK" STARS
                                 - 2) KINEMATICS ALSO USED TO CONSTRAIN MEMBERSHIP IN MOVING GROUPS
                                 - 3) KINEMATIC DISTANCES FEEDS INTO HRD
Data actions:
 Analysis actions:
I still need to calculate a few more group space motions
(for T associations with no published UVWs) before calculating "final"
membership probabilities for FEPS targets. I'm also trying to figure out
if/how to weight membership probabilities (what do you do when you have
a star that has a high membership probability to several groups? is a high
membership probability even useful at that point?).

General Concerns:


Assigning Final Ages:

figure illustrating correlations between different age diagnostics

 Discussion reveals that we are coming to an agreed procedure, once all of the details above are sorted out.

  • LAH suggested prioritization:
     
    1. cluster age
    2. HRD if inferred age is <20 Myr (probably want to enforce confirmation with one of the activity indicators) 
    3. R'HK     ACTION:  add switch to implement R'HK only if vsini < 15km/s and R'HK < -4.4
    4. Lx
    5. Li   ACTION: switch Li and Lx in prioritization?
    6. vsini 
          

    figure illustrating new vs current fiducial feps ages

    The error in age is set as the dispersion in the above, excluding cluster age.  
    figure characterizing the dispersion

  • The stars with the largest dispersion between the different indicators are:

    figure illustrating histograms of "best" and "averaged" ages. the former is defined according to the hierarchical list above. the latter is an average all of the non-cluster indicators counting the HRD as valid only for ages <20 Myr.

    figure comparing "best" and "average" ages directly

    Resulting number of stars with "best" age from each category. This demonstrates the rapid fade in applicability of age indicators towards the bottom of the list in the hierarchical age assignment technique.

    1. cluster age -- 113 stars
    2. HRD -- 41 stars with 3-5 obvious bogeys
    3. R'HK -- 129 stars
    4. Lx -- 63 stars
    5. Li -- 2 stars
    6. vsini -- 2 stars

    figure showing how well we would do by adopting the "average" ages for the cluster stars. this is surprisingly good at young ages and disturbingly bad at slightly older ages

  • Stauffer suggested procedure (LAH edited text):  Our immediate goal is to identify a path to assigning ages that we can reasonably implement and complete within the next few months with plausible level of effort from the ages group.

    Organization: below is a pathway to follow in assigning ages. More or less, the idea is to assign stars to one of several bins, and derive their age according to the bin they fall into. This may or may not address all of our stars - the idea would be to work through the proposed process, and see if all the stars are covered or not. If not, the process can be amended. The description below is schematic, and still needs to have each method "definitized" (e.g. whose Rhk' age relation, or derive our own).

    1) Use an agreed upon relation between CaHK normalized flux and age if have CaHK data and if: - vsini < 15 km/s - displacement above MS is either < 0.5 mag or unknown - not in a cluster or in an EEM moving group - Rhk' < -4.4 (all criteria to exclude stars that are too young for this method)    These criteria are intended to be both practical in the sense of keeping us in regimes where the data are good, and scientifically robust in the sense of keeping us away from activity saturation effects.

    2) Age from cluster or from an EEM approved moving group

    3) If have a distance, and if the displacement above the MS > 0.9 mag, then use isochrone age if not in (2).

    4) If neither (1) or (2), and have lithium data, derive lithium age from EEM algorithm. If also have age from (3), use average. If lithium age is contradicted by (3), only use (3). If lithium age implies a displacement above the MS > 0.5 mag and that is not true, do not use lithium age.

    5) If neither (1) or (2), and have Lx/Lbol, derive xray age from a to-be-derived EEM algorithm (a la his lithium algorithm), using xray data for open clusters from Sofia Randich papers and also xray data from Sco Cen. If ages from (3) and (4) are also available, derive an average. If the age from Lx is contradicted by (3), do not use the xray age. If the xray age implies a displacement above the MS > 0.5 mag and that is not true, do not use the xray age.    Possible to combine coronal and chromospheric indicators, i.e. merge #1 and #5.


  • Mamajek suggested procedure: use medians of applicable ages.