Some progress has been made to add the marshall functionality & changes required for PUNTS. The major backend change made was to open up the association radius for nuclear transients in the contextual classifier from 0.5 arcsec to 1.5 arcsec (transient to host core separation). This is what produces the predicted classification you can see on the marshall tickets:

A couple of important caveats to note about the contextual classifier:

• transients associated with faint (~> 20.0 mag) catalogued host galaxies will not be flagged as nuclear transients. Many of these hosts will be faint compact galaxies and also for most catalogues the star-galaxy separation accuracy degrades at this limit.
• transients occurring near the core of galaxies known to host an AGN (Veron and Milliquas catalogues) will always be flagged as AGN not NTs.

On the frontend a filter dropdown has been:

The filter works in whatever sidebar list you visit, so if you’re in the inbox and select ‘predicted NT’ it will filter the inbox to show you only transients predicted to be a nuclear transient.

‘contextual classification’ and ‘association separation’ have also been added to the sort options:

For members of the TAT the likely workflow for prioritising NT classifications would involve:

1. opening up the inbox
2. filtering by NT
3. sorting by host core separation
4. scanning the filtered & sorted list for transients that meet the host redshift, magnitude and transient-host delta mag criteria and prioritising them for classification.

Another seldom used view of the marshall is the table view where you can see lists of transients in a table instead of tickets (hover over the eye button and select table). For example here’s the list of all NT in the marshall.

Click on the table column header to change the sort parameter/direction and click on the table row to view the transient ticket.

To pull out some stats and give you an example dataset, I’ve taken ATLAS and GAIA transients discovered in the past 3 months. Here are the numbers within 1.5'' of their host cores:

See first point in the caveats to the contextual classifier for the differences between the NT and SN annotations at the cores of galaxies.

Here are these populations plotted as a cumulative distribution function of separation from the host core. First unfiltered:

And again now filtering by those transients associated with galaxies with distance measurements:

Finally here’s the list of those 54 SN/NTs with host distances and within 1.5'' of the core: