Examples of outlying RVs #961
Comments
|
Thank you for this, @Emmavt ! @howardisaacson, @bjfultn, @shalverson, @lukehandley -- let's try to categorize these outlier RVs and make sure that we're working toward solutions for all of them. |
|
I would welcome examples of other outlier RVs from other users to be posted to this thread. |
|
Here's an example of 3 consecutive exposures of the same target (201092) which seem to have identical WLS attributes, but one of the three has ~km/s difference from the others, as well as double the SNR: KP.20240526.53616.48 |
|
KP.20240526.53833.94 has different TARG header keywords. |
|
@***@***.*** ***@***.***> The 202092 outlier may
be due to the header keyword: TARGRADV having an incorrect value. It is
zero, instead of -64.2 km/s, like it should be. It's strange that it has a
different value from the sequential observations, but we should correct
that first and see if it is still an outlier.
…On Tue, Nov 12, 2024 at 12:00 PM lukehandley ***@***.***> wrote:
Here's an example of 3 consecutive exposures of the same target (201092)
which seem to have identical WLS attributes, but one of the three has ~km/s
difference from the others, as well as double the SNR:
KP.20240526.53616.48
KP.20240526.53725.34
KP.20240526.53833.94 (Outlier)
—
Reply to this email directly, view it on GitHub
<#961 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAKFHJ6BPPN6PJ7K5PDPXXT2AJM6LAVCNFSM6AAAAABRU2CGVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZRGQ2TSNRRGI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
--
------------------------------------------------------------------------------------------------
Howard Isaacson
Research Scientist in Astronomy
University of California, Berkeley
------------------------------------------------------------------------------------------------
|
|
Looking through the history database, the radial velocity keyword was
zeroed out during the last exposure, 30 seconds before the end.
That should not matter, the values used are supposed to be those at the
start of the exposure, allowing the observer to reset the values before
read out begins.
It could be that the window for the beginning is too larger, I will ask.
Brad
On Tue, Nov 12, 2024 at 12:38 PM Howard Isaacson ***@***.***>
wrote:
… @***@***.*** ***@***.***> The 202092 outlier may
be due to the header keyword: TARGRADV having an incorrect value. It is
zero, instead of -64.2 km/s, like it should be. It's strange that it has a
different value from the sequential observations, but we should correct
that first and see if it is still an outlier.
On Tue, Nov 12, 2024 at 12:00 PM lukehandley ***@***.***>
wrote:
> Here's an example of 3 consecutive exposures of the same target (201092)
> which seem to have identical WLS attributes, but one of the three has
~km/s
> difference from the others, as well as double the SNR:
>
> KP.20240526.53616.48
> KP.20240526.53725.34
> KP.20240526.53833.94 (Outlier)
>
> —
> Reply to this email directly, view it on GitHub
> <
#961 (comment)>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AAKFHJ6BPPN6PJ7K5PDPXXT2AJM6LAVCNFSM6AAAAABRU2CGVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZRGQ2TSNRRGI>
> .
> You are receiving this because you were mentioned.Message ID:
> ***@***.***
> com>
>
--
------------------------------------------------------------------------------------------------
Howard Isaacson
Research Scientist in Astronomy
University of California, Berkeley
------------------------------------------------------------------------------------------------
—
Reply to this email directly, view it on GitHub
<#961 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAJJHGSAIIQ7Y22QN434EI32AJRLFAVCNFSM6AAAAABRU2CGVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZRGUZTIOJXGY>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***
.com>
|
|
Looking at the keywords from the L2 files, the outlier observation was processed as DCSNAME = 201091, i.e. the binary companion. Based on the observing notes, there was an interruption during observing these targets on that night which required restarting the sequence. Perhaps an observer incorrectly edited the OB? Edit: I'm seeing a similar issue for observations on another date, where two sequences were taken back to back (both with 'star' 201092) but the first sequence was processed as 201091. Outliers: Regular Sequence: This looks to be more of an accounting issue than a core processing issue. Edit 2: RVs on 2024-09-14 are affected by the WLS interpolation issue described by Emma. |
|
The outliers that I've encountered for my set of TESS follow-up targets: TOI-1184:
TOI-5726:
|
|
Howard and I went through and added comments about each of the observations listed here so far. In all cases we are either missing bracketing LFC calibrations and we need to wait until we implement an etalon-based drift correction, or the TARG header keywords set at the telescope from the OB are inconsistent. |
|
From Stephen Kane and I's 30 target program, I've attached another tranche of outliers I've found Particularly common problem dates appear to be anything taken between [2-24-2024,2-27-2024], [6-17-2024,6-18-2024], 9-8-2024, and 9-1-2024 |
Following up on meeting with Andrew and Howard yesterday, here are the observation IDs for several outlying RVs in my 2024A+B programs observing M dwarf multis. Several of them have WLS files that are separated by 2-4 days, which is likely the issue.
The text was updated successfully, but these errors were encountered: