Hi, it’s really cool to see LSST alerts in the brokers! I’ve been trying to navigate the broker ecosystem today and I was hoping to get some advice from broker developers/users (apologies if I’ve missed some obvious documentation). How can I use a broker to get a list of alerts that have an ssObjectId over the past N nights? Better yet, is there a broker that I can filter on sso related alert values such as orbital parameters?
Having looked at some broker documentation I would assume that an Antares filter would be the most likely choice. I saw that there was relevant sso filters/tags for ZTF data (sso_confirmed, sso_candidates) but there does not appear to be equivalent filters/tags available for LSST data? Is this something that will be provided as part of the solar system object activity community alert filter or does this need to be developed independently?
I also looked at Fink which already provides excellent SSO information provided you have a name/designation to query. Is there anyway to make use of the Fink API, in particular /api/v1/tags, to query for objects tagged as SSO (i.e. ssObjectId is not null?). All I could find on the existing Fink filters is in this repo but there does not appear to be anything for SSOs.
There is also SNAPS but there doesn’t seem to be a lot of information on a working website/API from the Rubin brokers page.
Now this is not real-time, nor API-based. In the spirit of what we do in ZTF, a LSST filter for SSO is coming. You’ll be able to stream data directly to you, and query the DB from the API as well. Stay tuned!
Thanks, Julien! That’s useful information. Would observations of SSOs that are not yet in the MPC database be classified as sso here? In other words, if one wanted to find in alert objects that might be SSOs but are not yet in any database, can this be done easily with Fink? I imagine that one would have to search for alerts where the template shows no star at that location and then try to link it with other similar alerts in this part of the sky.
No, you would have for example to download all singletons (alerts with no history – block b_is_new) over a few nights, and do the linking downstream (beware some singleton gets history in future nights).
Note that we have a linker for SSO candidates (so not yet reported to MPC) that should be integrated at some point (at we were doing for ZTF). From this point in time, you’ll be able to ask to “SSO candidates” (according to Fink, so not perfect ).
Hi James! In ALeRCE we have developed jupyter notebooks to interact with the data (both tables and image cutouts) stored by our broker, using python. In particular, you can try this notebook for querying both 1) known Solar System objects (through their ssObjectId or unpacked designation), or 2) candidate moving objects based on our stamp classifier (still in beta mode, feedback very welcome!) plus visual inspection. It includes queries of ssObjects that had their last detection over the past N nights, so it may be useful for the example you mention. Orbital parameters are not indexed in our database, but once you get a list of ssObjectIds, you can loop over their properties and select further subsets of objects/detections.
Note that our Explorer web interface shows diaObjects but not yet ssObjects, but that will be solved within some weeks. We’re actively working in improving our services and documentation/tutorials, so in case of questions/feedback please let us know through this forum!
Thanks @JulienPeloton, now that I have signed up for a fink service subscription I’ll try out the Data Transfer method.
And thank you @ale_munozarancibia, I’ve had a quick read through the notebook and it looks like it will be really useful. I will have a go at running it and see if I can tweak it for what I want to do!
Sounds great @jrob93! Please let me know if you find any issues. Today we made some changes to the client and repository, so that image cutouts are better displayed on github, and notebooks can run in Google Colab
).