TNS-reported FRB missing from CHIME catalog

[0xCoto]

The TNS has a CHIME FRB event reported under the name of FRB 20190224F. However, I could not manage to trace this FRB back to the CHIME catalog, nor the repeaters page.

Could you please give some input as to where exactly this event could have been derived from, and whether a different name has been assigned to this event on the CHIME catalog?

[shinybrar]

@0xCoto How are you retrieving the data, i.e.

Analyzing the CSV directly from chime-frb.ca or using the python package cfod. e.g.

from cfod import catalog
data = catalog.as_dict()

I am asking, because one of the data transformations from the originally provided csv and fits file could have a bug as well.

[0xCoto]

@shinybrar I downloaded the CSV dataset from the 'Catalog 1 as a csv file' button (http://chime-frb.ca/catalog), and noticed I had a TNS CHIME FRB on my older database which was not in the downloaded CSV. I also tried looking up the FRB via the web platform, and it was also missing.

[shinybrar]

@0xCoto We will look into this and report back to you @0xCoto . Given its the weekend, it could be sometime before we get a reply.

@andrewzwaniga Can you take point on this issue, since there seems to be a TNS name for the FRB and not a corresponding burst in the catalog.

[0xCoto]

Also missing:

• FRB 20181120A
• FRB 20190302B
• FRB 20190516A
• FRB 20190604B
• FRB 20190626A

There's a bunch more I've spotted, but I hope these examples will suffice!

[0xCoto]

On the topic of TNS entries, I'd like to ask for a clarification regarding the identity of the following FRB(s):

• FRB 20180908A (ref.: https://www.wis-tns.org/object/20180908a)
• FRB 20180908B (ref.: http://chime-frb.ca/catalog/FRB20180908B)

The CHIME catalog does not appear to contain 20180908A, but the reported S:N is slightly different between the two bursts (10.4 vs. 11.3). There are no reported values for the flux, width and fluence on the TNS, but I assume it is the same burst with the S:N simply revised after further analysis (later on, after the submission to the TNS).

They also appear to have the same reported burst time, but the lack of a (nearly) duplicate event here indicates it is probably a TNS duplicate event.

[andrewzwaniga]

Hi @0xCoto, thanks for your diligence. I am getting up to speed with this today and doing a deep dive. I will post an update a little later today.

[andrewzwaniga]

Hi @0xCoto, thanks for your patience.

I looked at the 7 events you reported so far. Please see the attached .csv file for a table summarizing my findings. I am planning to address this in the foremost by adding a comment to the TNS Object page for each candidate, summarizing the explanations I gave for each one.

Across the events you identified, there are three mutually exclusive issues going on here.

1. A Candidate had no intensity data so it was not included in the Catalog; but an entry was made on the public repeaters page (https://www.chime-frb.ca/repeaters) and the real-time meta data was submitted to the TNS for naming.
2. Same as 1. except there is intensity data.
3. Ungrouped events representing the same FRB Candidate were separately submitted to the TNS, creating "duplicates".

Only one event, 20190626A, falls under case 2.

Please let me know if you find or have found other instances like these and I will do the same to track down their explanations.
chime-frb-open-data_community_issues_7 resolution - Explanations.csv

[shinybrar]

Adding the FRB's here for more visibility.

TNS Name	CHIME/FRB Event ID	Explanation
20190224F	31626218	R2 (20180814A) burst that is not published anywhere due to no intensity data captured. But it appears on the public repeaters page.
20181120A	20887861	The TNS object page for this burst says that this is published in https://arxiv.org/abs/1908.03507. It is a duplicate of 20887865 (20181120B).
20190302B	32227968	R5 (20181128A) burst that is not published anywhere (no intensity data); but it appears on the public repeaters page.
20190516A	38833513	R14 (20190212A) burst published in Table 2 in RN2 paper (R14 was published as source #3). NO CALLBACK DATA
20190604B	40363753	The TNS object page says that this is published in https://arxiv.org/abs/2001.10275. It is a duplicate of 40363754 (20190604F)
20190626A	42545361	R2 (20180814A) burst that is not published anywhere because it was detected after the R2 paper; but it appears on the public repeaters page.
20180908A	12806027	R15 (20180908A) burst that is ungrouped to 12806028 (20180908B). These are duplicates of the same event.

@andrewzwaniga What is the due course of action here?

• I think especially for the event with the intensity data, we should publish catalog v1.1

[andrewzwaniga]

Regarding the TNS, I think the action item there is:

• leave comments on the TNS object pages, indicating lack of intensity data capture or that the event is a duplicate of some other. (e.g. I just updated the "Remarks" in the "Photometry" table for 20190224f https://www.wis-tns.org/object/20190224f).

Regarding the CHIME/FRB Catalog 1, especially the burst with intensity data that did not appear in the Catalog, we have two action items:

• determine what led to the exclusion from the catalog
• confer internally in CHIME/FRB to evaluate the case for releasing a catalog v1.1.

[0xCoto]

Hi @andrewzwaniga,

Does the TNS have some kind of changelog for this page? I believe my database is missing one FRB that the TNS has, and I'm trying to see which one it could be, but I don't quite remember what the total number of events were on the TNS last time I checked.

[andrewzwaniga]

I am not aware of any official change log for that page. The closest I can think of is the daily emails that TNS sends out with newly reported FRBs around 22:00 UTC, for those registered for that. But admittedly that's not a programmatic way of assessing changes.

At this stage it may amount to downloading the full CSV of TNS search results and checking against the Discovery Date (UT) column therein.

[0xCoto]

So, I did manage to spot one new ASKAP event (FRB 20210404B) which was just published to the TNS, and ended up tracing the original FRB 20180908A that was missing from my database (which is also only part of the TNS and not the CHIME catalog).

No idea why exactly this outlier could have somehow been the only one left out, because it doesn't appear very new (seemingly reported in 2020). I checked and every other event between FRBSTATS and the TNS matches, so that seems odd. But in any case, I've added this event to the FRBSTATS platform by hand, so that everything is accurate and up-to-date.

---

Please let me know if you find or have found other instances like these and I will do the same to track down their explanations.

I did some data parsing and prepared a list. These are the only instances missing from the CHIME catalog & present in the TNS:

• FRB 20181120A
• FRB 20190224F
• FRB 20190302B
• FRB 20190516A
• FRB 20190604B
• FRB 20190626A
• FRB 20190702A
• FRB 20190702B
• FRB 20190718A
• FRB 20190806A
• FRB 20190809A
• FRB 20190810A
• FRB 20190810B
• FRB 20190815A
• FRB 20190817A
• FRB 20190824A
• FRB 20190825A
• FRB 20190825B
• FRB 20190825C
• FRB 20190907A
• FRB 20190915A
• FRB 20190925A
• FRB 20190926A
• FRB 20190929A
• FRB 20191013A
• FRB 20191018A
• FRB 20191020A
• FRB 20191021A
• FRB 20191029A
• FRB 20191029B
• FRB 20191030A
• FRB 20191030B
• FRB 20191107A
• FRB 20191110A
• FRB 20191111A
• FRB 20191113A
• FRB 20191116A
• FRB 20191117A
• FRB 20191202A
• FRB 20191205A
• FRB 20191215A
• FRB 20191217A
• FRB 20191218A
• FRB 20191219A
• FRB 20191219B
• FRB 20191219F
• FRB 20191220A
• FRB 20191223A
• FRB 20191231A
• FRB 20200112A
• FRB 20200120A
• FRB 20200120E
• FRB 20200122A
• FRB 20200122B
• FRB 20200122C
• FRB 20200122D
• FRB 20200122E
• FRB 20200122F
• FRB 20200122G
• FRB 20200124A
• FRB 20200124B
• FRB 20200203A
• FRB 20200204A
• FRB 20200204B
• FRB 20200204C
• FRB 20200204D
• FRB 20200204E
• FRB 20200219A
• FRB 20200220A
• FRB 20200221A
• FRB 20200224A
• FRB 20200302A
• FRB 20200307A
• FRB 20200314A
• FRB 20200324A
• FRB 20200718A
• FRB 20201124A
• FRB 20201124B
• FRB 20201129A
• FRB 20201210A
• FRB 20201216A
• FRB 20210220A
• FRB 20210301A
• FRB 20210321A
• FRB 20210322A
• FRB 20210323A
• FRB 20210326A
• FRB 20210327A
• FRB 20210327B
• FRB 20210327C
• FRB 20210328A

[shinybrar]

@0xCoto The CHIME/FRB catalog only contains events detected between July 25 2018 to July 1 2019. Most of these additional events are part of other papers from the CHIME/FRB Collaboration, and will be included in Catalog 2.

[0xCoto]

@andrewzwaniga - Is it possible to download the entire CSV from the TNS page? I typically have to:

1. Set Results in page to 500
2. Click Download as CSV
3. Go to next page (because CSV only contains the first 500 results)
4. Click Download as CSV
5. Merge the two CSVs together offline

If Results in page had a limit of e.g. 10000, that would be quite useful. Otherwise, if there's an easy way to achieve this over an API, that would be nice as well.

[andrewzwaniga]

Hi @0xCoto, in terms of FRBs on the TNS, the method you describe above is the best I know of. I agree with you that downloading them all in one shot is desirable, especially if it can be done through API.

It may be possible to suggest this to the developers on the TNS side. I can write to them and get back to you.

[0xCoto]

@andrewzwaniga - That'd be great if you could do that, thanks!

[andrewzwaniga]

I have written to developers of the TNS with ideas suggested by @0xCoto . I can post an update if or when I have one.

[nidhisainicloud]

@0xCoto @andrewzwaniga Is there any method to download the FRB data from TNS along with columns Flux, Fluence, burst width, etc, as these are not listed in the "Columns to display"?

[0xCoto]

@nidhisainicloud - You can do this programmatically using something like frbcat. However, there seem to be some issues lately, so my fork might be a better option. I just submitted a pull request, so unless this is urgent, I'd suggest waiting a day or two for the developers to approve the PR: TRASAL/frbcat#15

To tackle all the issues with the TNS, I've worked on a similar platform that should allow you to obtain the data you need much more easily (all TNS entries should be part of FRBSTATS): https://www.herta-experiment.org/frbstats

More details can be found on the paper, but you can download the data you're interested in as a CSV, JSON, Google spreadsheet, or even query things programmatically using the API.