Astro-COLIBRI Citizen Science Program

• Astro-COLIBRI Citizen Science Program is a real-time multimessenger platform that enables both citizen scientists and professionals to receive, filter, and act on alerts for transient astronomical events.
• The program employs a robust technical architecture with a public REST API, real-time databases, mobile applications, and custom observability tools to streamline rapid follow-up planning.
• Community features like discussion forums, coordinated observing networks, and partner integrations foster collaborative decision-making and enhance the scientific impact of amateur contributions.

Astro-COLIBRI Citizen Science Program denotes the citizen-science layer of Astro-COLIBRI, the COincidence LIBrary for Real-time Inquiry for Multimessenger Astrophysics: a real-time platform that ingests, organizes, and visualizes alerts from gamma-ray bursts, fast radio bursts, optical transients such as novae and supernovae, gravitational-wave events, and high-energy neutrino events. In the literature, this layer first appears as explicit “Citizen Science Support” inside a platform that “serves both professional and amateur astronomers,” and later as a structured program spanning local astronomy clubs, national amateur networks such as RAPAS, and international initiatives including the International Astronomical Union Citizen Science Program and UNOOSA’s “Open Universe” initiative (Schüssler et al., 2024, Schüssler et al., 14 Jul 2025).

1. Historical emergence and conceptual scope

Astro-COLIBRI entered the literature as a real-time coordination and decision-support platform for the transient, multimessenger sky, with mobile push notifications, a REST API, a static and a real-time database, a cloud-based alert system, and website and mobile clients (Reichherzer et al., 2021). Its early scientific framing was operational rather than pedagogical: rapid follow-up of short-lived, rare, and unpredictable events in time-domain and multi-messenger astrophysics. Even at that stage, however, the architecture already exposed public interfaces and user-configurable preferences that made near-simultaneous access by non-professionals technically possible.

A recurring misconception is that Astro-COLIBRI is primarily a citizen-science classification project in the style of Zooniverse. The papers describe a different model. Astro-COLIBRI is “not a citizen-science ‘project’ in the sense of Zooniverse-style classification tasks,” but a real-time platform deliberately designed so that citizen scientists and amateur astronomers can participate directly in rapid-response multi-messenger astronomy (Reichherzer et al., 2022). Its citizen-science component is therefore infrastructural: alert ingestion, filtering, observability analysis, follow-up planning, and coordination.

By 2023 and 2024, the citizen-science framing became explicit. “Citizen Science Time Domain Astronomy with Astro-COLIBRI” isolated three amateur-oriented capabilities—notification streams for very bright events, custom observer locations, and optimized observation plans for gravitational-wave events—while “Astro-COLIBRI: Empowering Citizen Scientists in Time Domain Astronomy” described a discussion forum, a weekly Top-10 transient list, and strengthened collaboration with amateur networks (Schüssler et al., 2023, Schüssler et al., 2024). In the 2025 formulation, the Citizen Science Program is no longer an inferred extension but an articulated program “woven into the core design” and organized across grassroots, national, and global scales (Schüssler et al., 14 Jul 2025).

2. Technical architecture and access modalities

The program rests on the same technical stack as the broader Astro-COLIBRI platform: a public RESTful API, real-time databases, a cloud-based alert system, and user clients in the form of a website and mobile applications for iOS and Android (Schüssler et al., 2024). Later architectural descriptions add implementation detail: the website is built with the Flutter framework, the mobile apps use the same Flutter front end compiled as native apps, the back end uses Python and Flask, the main persistent store is MongoDB, the low-latency update layer is Firebase, and user notifications are distributed through Firebase Cloud Messaging (Avila et al., 2 Feb 2026).

Alert ingestion is heterogeneous by design. The back end continuously handles GCN JSON notices transmitted via Kafka, GCN circulars ingested with custom Python parsers, TNS notifications, VOEvent-formatted alerts processed via comet brokers or Kafka streams, and some email-based alerts. Incoming messages are translated into a unified JSON-based Astro-COLIBRI format; if a trigger ID already exists, the message is treated as an update to an existing event (Avila et al., 2 Feb 2026). This normalization layer is central to the citizen-science program because it shields amateur users from the syntactic heterogeneity of GCN circulars, VOEvent XML, broker-specific schemas, and manual reports.

The public-facing design is explicitly open. The platform provides a public website, smartphone applications, and a public RESTful API. In the 2026 architecture description, “the majority of endpoints are publicly accessible and fully documented” at www.astro-colibri.science/apidoc, while only database-modifying operations are restricted (Avila et al., 2 Feb 2026). Earlier work already emphasized that Astro-COLIBRI informs users in real time “via push notifications on their mobile phones,” with notification latencies of “a few 100 ms” after the event is processed by Astro-COLIBRI (Reichherzer et al., 2021).

The resulting access model is asymmetrical but coherent. Read-heavy operations—event browsing, cone searches, event searches, visibility plots, observability summaries, and notification subscriptions—are public or broadly accessible. Write-heavy operations remain controlled. This asymmetry explains why early citizen-science use emphasized alert reception, target selection, and follow-up planning rather than direct in-platform ingestion of amateur measurements.

3. Citizen-science-oriented alerting, filtering, and observability

The program’s operational core is selective real-time alerting. Astro-COLIBRI evaluates incoming messages in real time, filters them by user-specified criteria, and contextualizes them in their multi-wavelength and multi-messenger environment (Schüssler et al., 2024). For non-professional observers, the most consequential feature is not mere access to alerts but access to filtered alerts that are realistically observable.

The key amateur-facing notification streams were formalized in 2023. One stream delivers “Bright optical transients (mag < 18),” defined as classified optical transients—supernovae and other optical transients—with magnitude at the time of detection $m < 18$. A second, Unistellar-specific stream delivers “bright and early optical transients,” typically requiring $m < 16.3$ at detection (Schüssler et al., 2023). These thresholds operationalize telescope-class constraints: the program suppresses alerts that would be observationally inaccessible to typical small- and medium-aperture amateur systems.

Filtering is multi-level. Astro-COLIBRI exposes top-level filters by observing facility and event type, and detailed filters through submenus. The examples given in the literature include optical-transient magnitude at detection, high-energy-neutrino “signalness” and event type, and gravitational-wave parameters such as significance, localization uncertainty, and event-classification probabilities such as $P_{\text{BNS}}$ and $P_{\text{BBH}}$ (Schüssler et al., 2024). The purpose is not merely user convenience. It is a mechanism for matching the event stream to an observer’s practical capability and scientific interest.

Custom observatory definitions are equally central. Users can choose pre-defined professional observatories, query the International Astronomical Union observatory database through an auto-complete search, or define custom observatories either from device GPS or by manually entering coordinates and site parameters. A free user account created with an email address is required to save these custom sites (Schüssler et al., 2023). Once defined, the site propagates through all observability and scheduling tools.

The observability machinery is standard spherical astronomy. For an event at right ascension $\alpha$ and declination $\delta$, observed from latitude $\phi$ and hour angle $H$, Astro-COLIBRI uses the standard altitude relation

$$\sin h = \sin\phi\,\sin\delta + \cos\phi\,\cos\delta\,\cos H,$$

and airmass approximations such as

$X \approx \frac{1}{\cos z} = \frac{1}{\sin h},$

with additional Sun and Moon constraints for practical scheduling (Reichherzer et al., 2021). In the citizen-science context, these calculations appear not as raw ephemeris output but as visibility plots, observing-condition evaluations, and immediate assessments of whether a source is up, how high it gets, and when it is best observed.

A further specialization concerns gravitational-wave follow-up. Astro-COLIBRI integrates the public tilepy scheduler and exposes it through the visibility interface. The user selects an observatory, requests a follow-up schedule, and receives a list of pointings plus their ordering, displayed both in a table and as fields overlaid on the GW skymap (Schüssler et al., 2023). The associated optimization uses the GW localization probability map $m < 16.3$0, integrating probability over a field of view; in the formulation given for HEALPix-like tiling, a tile score is proportional to $m < 16.3$1 (Schüssler et al., 2023). This made probabilistic tiling of wide GW localizations available to observers who were never expected to manipulate HEALPix products directly.

4. Community structures, partner networks, and governance by collaboration

The citizen-science program is not limited to an individual user receiving alerts on a phone. It includes explicit community structures. A discussion forum at https://forum.astro-colibri.science functions as a coordination layer in which users discuss interesting events, share observing plans and results, and vote on promising targets (Schüssler et al., 2024). In collaboration with RAPAS, Astro-COLIBRI also produces a weekly Top-10 transient list posted in the forum; this list is intended as a curated observing program for amateurs and includes a voting system that helps steer community attention (Schüssler et al., 2024).

Two organized amateur networks are directly identified in the 2024 citizen-science overview. RAPAS (https://rapas.imcce.fr/) uses Astro-COLIBRI to compile the weekly Top-10 list and to support manual event selections and submissions. BHTOM (https://bh-tom2.astrolabs.pl/) is likewise linked through manual event selection and direct submission by network members (Schüssler et al., 2024). These integrations are important because they move Astro-COLIBRI beyond an alert dashboard toward a coordination substrate for existing Pro-Am networks.

The 2025 programmatic description arranges participation across three scales. At the grassroots level, the platform collaborates with local astronomy clubs and provides free, ad-free tools for transient monitoring. At the national level, networks such as RAPAS use Astro-COLIBRI’s real-time capabilities for coordinated observations. At the global level, the program participates in the IAU Citizen Science Program and UNOOSA’s “Open Universe” initiative, and it acts as an alert engine for the Unistellar Citizen Science Program (Schüssler et al., 14 Jul 2025).

The collaboration with Unistellar adds an automation layer. In the “Cosmic Cataclysms” workflow, ZTF discoveries enter brokers such as ALeRCE, the Unistellar science team curates a subset of bright early candidates, Astro-COLIBRI distributes these through a dedicated notification stream, and one tap in the app sends pointing commands to a Unistellar telescope; observations are then uploaded to a science server at the SETI Institute (Schüssler et al., 14 Jul 2025). The same paper describes an Astro-COLIBRI plugin for N.I.N.A. that can auto-populate the transient’s coordinates and configure the imaging sequence (Schüssler et al., 14 Jul 2025). These partner-specific integrations illustrate how the citizen-science program extends from passive notification to semi-automated observing.

5. Documented scientific use cases and amateur contributions

The citizen-science program is best understood through its documented observing cases. In “Astro-COLIBRI 2,” the platform is described as having become “an integral part of many amateur astronomers’ quest to study bright supernovae,” and two examples are given. SN 2022eyj, a Type Ia supernova at redshift $m < 16.3$2, was discovered by ASAS-SN on 22 March 2022 at 07:26:24 UT at RA $m < 16.3$3, Dec $m < 16.3$4, with brightness 16.1 mag at detection; the amateur user @SacHA(P) received an Astro-COLIBRI push notification and performed optical follow-up observations. SN 2022hrs, a bright Type Ia supernova at redshift $m < 16.3$5, was discovered on 16 April 2022 at 14:50:40 UT at RA $m < 16.3$6, Dec $m < 16.3$7, with magnitude 15 at detection; the amateur user @Stef_Astro likewise followed up after receiving an Astro-COLIBRI notification (Reichherzer et al., 2022).

A different form of citizen-science contribution appears in the SN 2023ixf case. The 2023 high-energy-oriented overview notes that amateur astronomers monitoring M101 detected this bright Type II supernova very early in its evolution, constraining the explosion time to about one hour. The same paper cites the Unistellar citizen network as having produced a densely sampled light curve with average sampling time 3.3 hours over 35 days (Schüssler et al., 2023). Here the scientific payoff was not limited to supernova photometry itself; the narrow explosion-time constraint strengthened high-energy neutrino searches by sharply reducing the temporal search window.

The 2025 citizen-science program description provides a further Pro-Am example through SN 2025coe, identified as a Type Ib, Calcium-rich supernova discovered by amateur astronomer Koichi Itagaki and reported to TNS. Astro-COLIBRI ingested the alert, distributed it through the dashboard and push-notification system, and RAPAS members coordinated rapid follow-up photometry. In the resulting light curve, RAPAS data in red, blue, and green were combined with ATLAS and ZTF measurements, yielding a densely sampled multi-band record (Schüssler et al., 14 Jul 2025). The significance of this example is methodological: the amateur contribution is not treated as anecdotal supplementation but as photometric data merged with professional survey products inside the same event representation.

These cases clarify the practical workflow envisioned throughout the program. A user receives a filtered alert, inspects the event view, checks observability from a custom site, decides whether the target is within instrumental reach, observes it manually or through integrated automation, and reports or shares the result either through partner infrastructures such as RAPAS or Unistellar or through the program’s community channels. The literature repeatedly presents amateur observers not as passive recipients of outreach material but as components of the observational network.

6. Misconceptions, limitations, and anticipated developments

A first limitation is conceptual. Astro-COLIBRI’s citizen-science layer has often been inferred from platform capabilities before being fully formalized. Early and mid-period papers explicitly state that they do not describe a complete citizen-science project with standardized upload pipelines or classification tasks; they instead describe a platform whose design enables citizen-science participation (Reichherzer et al., 2022, Schüssler et al., 2024). This distinction matters because some expected features of mature citizen-science systems—task assignment, contributor scoring, in-platform data validation, or publication workflows for amateur uploads—were initially absent or externalized.

A second limitation is observational heterogeneity. Amateur instruments vary widely in aperture, field of view, detector performance, and automation level. Astro-COLIBRI addresses this partly by allowing user-defined observatory locations and filters appropriate to a user’s own capabilities, and partly by using bright-event streams such as $m < 16.3$8 and $m < 16.3$9 to reduce mismatch between alerts and equipment (Schüssler et al., 2023). But the literature also identifies alert overload, diverse instrumentation, and the complexity of low-significance GW and neutrino alerts as persistent challenges (Schüssler et al., 2024).

A third limitation concerns data return. The short 2024 Moriond paper explicitly notes that detailed mechanisms for uploading amateur data or integrating it back into the system are not described there (Schüssler et al., 2024). Even in the 2025 programmatic description, amateur photometry often flows through partner infrastructures—RAPAS, Unistellar, or formal channels such as TNS AstroNotes—rather than through a universal Astro-COLIBRI upload protocol (Schüssler et al., 14 Jul 2025). The system therefore already supports community observation, but its feedback loop remains only partially internalized.

Future development has been framed along three lines. First, more collaboration tools: the 2021 architecture paper’s outlook mentions a comment function on individual events to “enable a better exchange of relevant information on specific transient events” (Reichherzer et al., 2021). Second, more automated interpretation: the same paper announces deep learning-based natural language processing of human-written alerts such as GCN circulars to extract updated localizations, classifications, and counterpart information (Reichherzer et al., 2021). Third, broader citizen-science integration: later papers mention additional broker integration, continued expansion of amateur-network collaboration, Astro-COLIBRI GPT, and the possibility of providing results of user-performed follow-up observations directly into Astro-COLIBRI (Schüssler et al., 2024, Reichherzer et al., 2022).

In that sense, the Astro-COLIBRI Citizen Science Program occupies an intermediate but technically distinctive position in contemporary astronomy. It is neither a classical amateur-observer mailing list nor a pure crowd-classification system. It is a real-time, multi-messenger coordination environment in which citizen scientists receive the same alert streams and much of the same decision-support context as professional observers, but through interfaces, filters, community structures, and partner integrations designed to make rapid follow-up by non-professionals scientifically actionable.