Using BASE-9 with Gaia filter

[thinhnh3009]

Hello,

Thank you very much for this great code! I am an undergraduate student and right now I am using BASE-9 for my research project about determining the age of open cluster NGC 2509 with Gaia EDR3 data. When using BASE-9, I edited the base9.yaml to use the PARSEC model, and I formatted my photometry as attached below (the photometry columns as "G", "G_BP", and "G_RP", matching with the PARSEC.model file). However, the code returned "Found unsupported star in photometry". Could anyone please have a look to see what I did wrong? Thank you very much!

data_member.zip

[tedvh]

Hi -

First, I'd like to know your name if you are going to be asking us for help. And while we are at it, who are you working with and at what university?

Second, you need to incorporate the photometric errors. These go right after the magnitude columns. So in your case what you have now

id G G_BP G_RP stage Cmprior useDBI

needs to become

id G G_BP G_RP sig_G sig_G_BP sig_G_RP mass1 massRatio stage Cmprior useDBI

or possibly

id G G_BPx G_RP sig_G sig_G_BPx sig_G_RP mass1 massRatio stage Cmprior useDBI

where x = "br" or "ft" depending on the Gaia BP filter brightness (that might be just for DR2, I cannot remember).

The above also includes your starting estimate of the mass of each star "mass1", which need not be that close, and if you think it is a binary, the mass ratio of that binary. You can enter 0.0 for massRatio and it should find any unresolved binary if it is really there and contributes meaningfully to the photometry. I would also suggest you set useDBI to 0 or 1, i.e. an integer, and not 1.0, though it might work fine with that.

-Ted von Hippel

[thinhnh3009]

Hello Dr. von Hippel,

I am very sorry for my mistake of not introducing myself. My name is Thinh Nguyen, and I am a senior undergraduate student at Villanova University. I am working on this project with Dr. Andrej Prsa and Dr. Scott Engle. I will try implementing your comments and let you know the results.

Thank you so much again. Thinh

[tedvh]

Thanks for introducing yourself Thinh and good luck with your work. We have a few videos from past workshops on the capabilities and use of the code. Perhaps these will be useful? https://www.dropbox.com/s/yxvoyh85xql5bbi/BASE-9%20Capabilities.mp4?dl=0 https://www.dropbox.com/s/4q13w0qko829b3a/BASE-9%20Use%2BUpgrading.mp4?dl=0

[thinhnh3009]

Hello Dr. von Hippel,

Thank you so much for the video! They are very helpful and I understand the code much better after watching them. However, I still have problems running the code. My input.phot file and .yaml file are attached below. Specifically, this is what the output looks like:

_Model boundaries are (7.420, 10.130) log years. Binaries are ON

***Warning: "G" is not available in the selected WD Atmosphere model This is non-fatal if you aren't using the WD models

Running Stage 1 burnin... Complete (acceptanceRatio = 0.000)

Running Stage 2 (adaptive) burnin... Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. Acceptance ratio: 0.000. Retrying. gsl: cholesky.c:157: ERROR: matrix must be positive definite Default GSL error handler invoked. Aborted (core dumped)_

Is there something wrong with my input file or yaml file? Thank you very much for your help!

PS: sorry it takes a long time to reply. I tried tweaking some minor things in the code and waited for it to run, but it didn't really work out.

NGC2509_4.zip

[tedvh]

The problem is you aren't using reasonable starting values or reasonable priors so the Markov chain cannot find a fit. These values are in the base9.yaml file. Your base9.yaml file has

---

cluster:
    starting:
        Fe_H:        0.07
        Av:          0.009
        Y:           0.29
        carbonicity: 0.38
        logAge:      9.0

        # You may have distMod or parallax, but not both
        #
        # The starting parameter, prior mean, and prior sigma must all
        #  specify the same parameter
        distMod:     0.0
        # parallax:    0.0

    priors: 
        means:
            Fe_H:        0.07
            Av:          0.009
            Y:           0.29
            carbonicity: 0.38
            logAge:      9.0

            distMod:     0.0
            # parallax:    0.0

        sigmas:
            Fe_H:        0.05
            Av:          0.006
            Y:           0.0
            carbonicity: 0.0   # .inf for uniform, 0 for no sampling
            logAge:      .inf  # .inf for uniform, 0 for no sampling

            distMod:     0.05
            # parallax:    0.05

---

the most unreasonable of these parameters is that the distance modulus (distMod) = 0.0. That is a cluster at 10 pc. Set the distance modulus starting value and prior to reasonable values and also set the [Fe/H] and Av values to be consistent with what is known for your cluster. Your advisors may have good values for these parameters.

For further discussion about your use of the code, let's switch over to using email. Email me at ted.vonhippel@erau.edu.

-Ted von Hippel