Alerts and brokers

Website: alerce.science

ALeRCE enhances the scientific return from Rubin alerts by providing a dynamic catalog of aggregated, cross-matched, annotated, and classified light curves. We prioritize delivering carefully curated machine learning classification probabilities across a broad taxonomy, supporting the time-domain community in the study of stochastic, transient, and periodic objects. ALeRCE also offers web interfaces for general object exploration, early supernova identification (including reporting to the Transient Name Server), and a watchlist system for monitoring objects of interest. Astronomers can easily access ALeRCE’s services through web portals, a Python client, APIs, or direct database queries, with access to the Rubin Science Platform. This enables rapid target prioritization, efficient follow-up, and deeper scientific analysis. By using ALeRCE, astronomers gain a comprehensive, scalable platform that accelerates discovery and drives impactful science in the time-domain era.

Website: ampelproject.github.io

Description to be added soon.

Website: antares.noirlab.edu

ANTARES is a full-service, real-time broker that adds contextual value to ingested alerts from multiwavelength astronomical catalogs (e.g, Gaia, Sloan, WISE, Chandra) as well as the past history of the event. Users can write their own filters in Python to identify specific classes of objects, create watch lists for direct notification, or develop catalogs for large-scale comparisons. Filters can range from something as simple as a magnitude limit to one that compares to a machine-learning based light-curve shape model. Users can interact with the system via a web portal, API, or substreams of alerts from filters.

Website: docs.babamul.dev

Babamul transforms the massive and heterogenous LSST data stream into smaller, focused channels organized by astronomical event properties (e.g. fast, blue, ...). Each channel delivers lightweight, science-ready alerts enriched with context from real-time surveys including ZTF. By dramatically reducing the data volume and splitting the stream in its different parts, you can subscribe only to events relevant to your research and filter them further using minimal hardware and familiar tools (e.g. astropy, numpy, astroquery) while also allowing for fast and local ML-based workflows. To lower the barrier of entry and jump-start projects, we also provide a number of open-sourced modules, ML models, example pipelines, and data management systems (with candidate vetting and follow-up UIs) inheriting from 7+ years of experience discovering transients in the ZTF alert stream.

Website: fink-broker.org

Fink is a community driven project that processes time-domain alert streams and connects them with follow-up facilities and science teams. To enrich and filter the data we use catalogue and stream cross-matches, feature engineering and Machine Learning algorithms. Fink was designed to be modular and to facilitate the inclusion of personalized science modules and filters. All tools already deployed can be accessed via our webportal, API or through a kafka stream. If you have a specific science case you would like to be developed, contact us at contact@fink-broker.org.

Website: lasair.lsst.ac.uk

Users make simple or complex alert filters with SQL, using webpage or API. A filter can run in real-time as alerts come in, with machine-readable output. Filters can be based on lightcurve features, intelligent crossmatch, and geometry. To get supernovae from Lasair, filter for high-temperature fast risers that have a host galaxy. The Sherlock intelligent crossmatch allows filters based on sky context: galaxy catalogues, known CV, variable star, etc. Users can work with ready-made public resources such as Gaia white dwarfs, galaxies within 5 Mpc, etc, then duplicate, modify, share, and annotate.

Website: pitt-broker.readthedocs.io

The Pitt-Google Alert Broker is a scalable, low-latency alert distribution service that will enable broad public access to, and scientific analysis of, LSST’s full alert stream with a low barrier to entry. Our broker’s cloud-native model uses Google Cloud Platform (GCP) to facilitate access to the data products we serve, and our Python library (pittgoogle-client) simplifies the way users can interact with our broker’s services and alert data. To learn more about how we can support your science, explore our demos page and open an issue if you have any questions.

Website: Trilling et al. (2023)

SNAPS -- the Solar System Notification Alert Processing System -- is a downstream broker that ingests Solar System observations from ZTF, LSST, and several other surveys. For each Solar System object we derive a number of properties (color, lightcurve, etc.), and serve both the collected observations and these derived properties. We search for both individual and population outliers, and will issue alerts about those unusual objects. The community can interact with our database through our web portal, where users can designate their "favorite" objects, and (coming soon) an API.

Website: poibroker.uantof.cl

Point of Interest is a downstream broker from ANTARES. It receives variable star alerts and carries out feature calculation, while keeping and analyzing the feature history. In this way, it is well suited for tracking the behavior of variable stars and aiding especially in the detection of RR Lyrae with Blazhko effect as well as anomalies.