herta / frbstats Goto Github PK

Could be easily achieved with pygedm, and added as a new column.

.htaccess has to be modified accordingly to strip trailing slashes from directory URLs. This currently only works with .html and .php files.

I really love the tree view of FRBs but it is not as insightful as cool it looks.

Maybe, it would be super cool if each repeater FRB has it's own burst v/s time plot.

There are atleast two known FRBs which have a periodicity (FRB20121102A, FRB20180916B). Would it possible to show the periodicity in action? A plot like Fig.3. of Marthi etal 2020.
I can put in the math here if you'd like.

There is an FRB2021 conference happening sometime in July online. Here's the advt
Although I don't know what the committee would decide, but if you haven't considered already, I urge you to submit an abstract/take-part/advertise in the conference! It would be really cool.

lastly, dark mode please

thanks

The following blocks need to be updating dynamically upon new catalogue entries:

• Events observed
• FRB repeaters
• Galactic sources
• Event coverage
• FRB count vs MJD
• FRB class

To-do: sitemap submission.

Cluster analysis algorithms can be evaluated based on the current repeaters.json dataset (assuming 100% of the entries are true, which is most likely the case). This can dramatically skyrocket the autonomy of FRBSTATS and significantly reduce manual maintenance.

Classification parameters to be considered are probably l, b and DM. Everything else (other than RA/Dec. of course) may be dependant on the emission mechanism of the burst and/or detection instrument (telescope), which can introduce confusion.

Maybe l, b errors can be dynamically accepted based on the beamwidth of the telescope that observed the burst. That will hopefully not be necessary.

These two need to be checked, because they don't appear to be detected by DBSCAN:

• https://www.chime-frb.ca/repeaters/190907.J08+46
• https://www.chime-frb.ca/repeaters/FRB20171019A

The FRB profile can be quickly displayed, with the help of either an onclick or onmouseover event.

Make it possbile to get the genereal stats (like total observerd events, repeaters, galactic sources, event coverage)
Maybe also accsess to the FRB Count over Time Graph, or the FRB Class graph (as png?)
Maybe also a filter option not to only search by name?!

Slight row color alternations helps navigate through the data more conveniently.

Something like this could work:

table.dataTable td.sorting_1 { background-color: white; border:1px lightgrey; }
table.dataTable td { background-color: white;  border:1px lightgrey; }
table.dataTable tr.odd { background-color: white;  border:1px lightgrey; }
table.dataTable tr.even { background-color: white; border:1px lightgrey; }

Thank you for your efforts in providing visual FRB data.
I found that some data in the Catalogue needs to be checked again.
The flux of FRB 20190614D should be 0.124 Jy. @0xCoto

The FRBSTATS catalogue.csv contains 15 parameters (columns) per FRB, most of which are currently empty (cells denoted with x) and need to be filled. Some columns are directly derivable from the current dataset (since there is an injective transformation function for certain parameters):

• utc → mjd
• (ra, dec) → (l, b)¹
• telescope & frequency → sefd (? - this is only an approximation, ignoring time-wise T_sys variability)

The rest shall be obtained from the relevant publications. Additionally, verification is needed for the present data derived from TNS. In particular, the flux column is highly unreliable and requires careful examination, including the use of the correct unit (i.e. Jy and not mJy).

Columns filled:

• FRB
• UTC
• MJD
• Telescope
• SEFD
• RA
• Dec.
• l
• b
• Frequency
• DM
• Flux*
• Width*
• Fluence
• S:N
• Reference

*Severely underpopulated due to lacking TNS data.

1 This transformation can be very easily carried out with the help of the Astropy or Virgo package. ↩

from virgo import galactic

# Convert RA/Dec. to Gal. Long./Lat. The position is returned as a tuple: (l, b)
galactic(ra, dec)

Current parameter columns

• FRB
• MJD
• UTC
• Telescope
• Right ascension (α)¹
• Declination (δ)¹
• Galactic longitude (l)¹
• Galactic latitude (b)¹
• Center frequency (MHz)
• Dispersion measure (pc cm⁻³)
• Peak flux density (Jy)
• Pulse width² (ms)
• Fluence (Jy ms)
• Verification status/reference

1 Most probable position (generally determined by the telescope beam). ↩
2 Generally the FWHM of a single-Gaussian fit to the FRB profile. ↩

The TNS contain an instance of FRB 20181130B, but no instance of FRB 20181130A. This may be an automatic error, because reporters are new to the designation (suffix letter format), and because the reported UTC time is 00:00:00, which is clearly inaccurate. Table 1 of the corresponding publication reports the exact MJD.

Same issues apply for FRB 20181118: while instances of FRB 20181118C and FRB 20181118D are present, no instances of FRB 20181118A or FRB 20181118B have been published to the TNS database.

(The TNS has published incorrect units for the flux density of these burst (mJy instead of Jy), but that's another issue.)

Flux and MJD corrections have been applied to the FRBSTATS catalogue and are visible on the online database.

There is no longer any content visible to populate the plots on the front page, and the repeater page is blank as well. Many of the data columns on the catalogue page are also no longer populating. I'm not sure if this is indicative of an issue on the frontend or in trying to query the existing database.

Add Save as PDF button to the /plots page and ensure all .py files export a .pdf as well.

The favicon currently displays the HeRTA logo, but it would be prefferable if it showed the FRBSTATS logo (DB icon).

An alert system shall be developed in order to notify subscribers of new FRB events via email, upon submission/disclosure to FRBSTATS/TNS. This is of high importance, as it permits astronomers to rapidly carry out critical (multi-wavelength) follow-up observations of FRB candidates.

Refer to similar FRBs from the repeater time series plots.

The implementation of a Sankey diagram allows visualizations involving the relationship between primary bursts and repeater events (parent → child).

Potential libraries:

• chartjs-chart-sankey
• Google Charts

Obtain FRB parameter data from the Transient Name Server (TNS) and FRBCAT (Petroff et al., 2016). This can easily be done (and automated for bridging new alerts/unconfirmed events from TNS automatically) with the Download as CSV functionality and a simple Python script for parsing the data to a database-compatible format.

The dedicated Object Search API may also be used to obtain data in a more rapid and repeatable manner.

While there does not seem to be a well-defined standard for the precise FRB designation, the format that appears prevail in the literature is FRB YYYYMMDDabc.

Furthermore, it may be a good idea to strip the trailing A from non-repeaters that is present in FRB events on TNS.

Many recent (and old) FRBs are reported to The Astronomer's Telegram. Enabling the Fast Radio Burst filter is useful for checking every event against the FRBSTATS database and ensuring everything is entered in, as there is a strong suspicion that not all ATel FRB posts end up getting pushed to TNS, and they may eventually get lost.

Plots

• Flexible plots (e.g. Gal. latitude vs DM); leave it to the user to judge whether a parameter combination makes sense
• Interactive plots

Catalogue rows

• Redshift (progress: redshifts.py)
• Localization area
• RM & polarization properties (frac. linear and circular)
• Scattering (temporal broadening and scint. bandwidth)

Database

• FRB widths with 1e3-1e4 ms are dubious (duplicate of #23)

Misc

• # FRB Source instead of FRB Count
• Association with catalogued sources within errorbox (galaxies, supernovae, X-ray sources etc.) - see: http://cdsxmatch.u-strasbg.fr
• Cross-reference positions and estimated redshifts of FRBs with the Nasa Extragalactic Database and find objects which could be potential sources

Repeater Tree not accessible，https://www.herta-experiment.org/frbstats/repeaters

The coordinates of an event shall be listed with their corresponding units on the Web platform.

RA, Dec., Galactic Longitude and Latitude can be displayed like so:

This can be applied to a web-oriented version of the CSV file, allowing the user to download the data as-is for further analysis locally.

Suggestion from Jonathan Katz: ability to print entire catalogue as a hard copy. Maybe a button would be accessible enough.

It would be nice if each repeater time series plot came with a small table containing the reported properties of each burst.

The identification of potential use cases for data plots provides valuable insights concerning the technical challenges behind graphical representations of FRB parameters.

Potential plots include, but are not limited to:

Line plot

• Total FRB detection count vs MJD

Pie chart

• Repeaters vs one-off events

(Stacked) bar chart

• Event count vs Telescope vs Year

Scatter plot

• Peak flux density vs Pulse width
• • DM may also be highlighted with a colormap on the scatter points
• Fluence vs DM
• RA vs Dec.
• Gal. Long. vs Gal. Lat.
• Fluence vs Center frequency
• S:N vs SEFD

Class plots may also be used to highlight the detection instrument on scatter plots.

Open question (TBD): error bars?

Some events appear to have exceptionally wide burst widths. Most of these appear to be associated with Pushchino LPA, so these events have to be double-checked and revised according to the relevant publications.

In order to ensure maximum database reliability for population studies, if such observation data fail to appear reliable, all corresponding rows shall be dropped from catalogue.csv.

Checked Pushchino events

• FRB 20160920A
• FRB 20170606A
• FRB 20151018A
• FRB 20180321A
• FRB 20160206A
• FRB 20141216A
• FRB 20161202A
• FRB 20131030A
• FRB 20140212A
• FRB 20121029A

While /FRBSTATS is recognized, /frbstats is not (case-sensitive URL). This should be fixed in order to help accessibility.

The current DB infrastructure handles json files better, so a parser is required to act like an interface between (TNS-obtained) csv-formatted files and MongoDB.

https://www.herta-experiment.org/frbstats/repeaters

An API needs to be built in order to allow developers to conveniently fetch, update, plot and experiment with database parameters.