🔭 astrophysics

BOCOSUR: An all sky network for fireball detection in Uruguay

This paper presents the deployment and technical evaluation of the Bocosur all-sky network in Uruguay, which consists of 20 autonomous stations designed to detect asteroidal fireballs for meteorite recovery and to engage secondary students in citizen science, while also validating a new photometric methodology and reporting initial results on a bright fireball detection.

Original authors: M. Caldas, A. Guaimare, V. Abraham, L. Barrios, M. Hernández, L. Velasco, G. Tancredi

Published 2026-05-12

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: M. Caldas, A. Guaimare, V. Abraham, L. Barrios, M. Hernández, L. Velasco, G. Tancredi

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the night sky as a giant, dark ocean. Occasionally, massive rocks from space (asteroids) dive into this ocean, burning up as they hit the atmosphere. These burning rocks create spectacular, bright streaks of light called fireballs. Sometimes, if the rock is big enough, a piece of it survives the plunge and lands on Earth as a meteorite.

Finding these fallen rocks is like trying to find a specific needle in a haystack, but the needle is moving at thousands of miles per hour and landing in a vast, dark field. To catch them, scientists need a team of observers watching the whole sky at once, all taking notes at the exact same moment.

This paper introduces BOCOSUR, a new "team of sky-watchers" located in Uruguay, South America. Here is how they built it and what they found, explained simply:

1. The "Sky Net"

Think of BOCOSUR as a giant safety net made of 20 cameras spread across the country.

The Locations: Instead of putting these cameras in remote deserts or on top of lonely mountains, the team placed them on the roofs of high schools.
The Human Element: This is a "citizen science" project. High school teachers and students are the guardians of these cameras. They help keep the equipment running and even help sort through the video footage. It's like turning a classroom into a mini-observatory.
The Goal: The net is designed to catch bright fireballs (the "needles") so scientists can calculate exactly where they came from and where they landed, making it possible to go dig them up.

2. The Upgrade: From "Old Glasses" to "Super-Vision"

For a while, the network used older, standard security cameras (called Watec). These were like wearing thick, foggy glasses; they could see the big picture, but the details were blurry.

The Change: In 2022, they swapped these out for high-definition digital cameras (the ZWO ASI 178MM).
The Result: This is like switching from foggy glasses to a pair of high-powered binoculars. The new cameras see much sharper details and can spot fainter stars.
Why it matters: Because the new cameras are so sharp, they can use the stars visible in the same short video clip of a fireball to measure the fireball's brightness. With the old cameras, they often needed a separate, longer video just to measure the stars, which created a massive amount of data to store and manage. The new system is faster and smarter.

3. How They Measure the Light

When a fireball flashes, it's often so bright it "blinds" the camera (saturates the sensor), making it hard to tell exactly how bright it really was.

The Analogy: Imagine trying to measure the brightness of a blinding flashlight by looking at a dim lightbulb nearby.
The Method: The team developed a mathematical trick. They use the known brightness of stars (and even planets like Jupiter and the Full Moon) as a "ruler." They compare the fireball to these known rulers to estimate the fireball's true brightness, even if the camera was overwhelmed by the light.
The Test: They tested this ruler against Jupiter and the Moon. It worked well, with only a small margin of error (about the difference between a slightly dim and slightly bright lightbulb).

4. The First Big Catch

The team put their system to the test with a very bright fireball spotted on October 29, 2022.

The Event: Five different cameras in the southeast of Uruguay saw the same fireball at the exact same time.
The Reconstruction: By combining the views from these five "eyes," they could build a 3D movie of the rock's path. They calculated:
- Where it started: About 100 km high in the sky.
- Where it ended: About 53 km high (it didn't reach the ground, so no meteorite was recovered from this specific event).
- How fast: It was traveling at about 32 kilometers per second (roughly 72,000 mph).
- Where it came from: They traced its path back to a specific group of space rocks known as the Southern Taurids, which are linked to a comet.

5. What's Next?

The network is now fully running. The team is working on teaching a computer to automatically sort through the thousands of hours of video to find fireballs without needing humans to watch every second. They also plan to use wind data to predict where meteorites might fall if they do survive the trip to the ground.

In summary: BOCOSUR is a high-tech, community-powered sky net in Uruguay. It uses sharp new cameras and clever math to track space rocks, helping scientists understand where they come from and potentially finding new meteorites to study.

Technical Summary of "BOCOSUR: An all sky network for fireball detection in Uruguay"

Problem Statement
Meteorites provide critical data regarding the formation and evolution of the Solar System, yet the number of meteorites with computed pre-atmospheric orbits remains extremely low (47 as of December 2023) compared to the total number of recovered meteorites (>70,000). A significant portion of these recovered falls are located in the Northern Hemisphere, with only two found in Latin America. To address this geographical bias and increase the recovery rate of meteorites, particularly those of asteroidal origin, the deployment of automated fireball networks is required. These networks must be capable of simultaneous multi-station detection to triangulate trajectories and define strewn fields. However, existing networks are sparse in the Southern Hemisphere, and many lack the specific infrastructure or citizen-science integration necessary for sustainable operation in regions like South America.

Methodology
The BOCOSUR project addresses these gaps by deploying a nationwide network of 20 automated all-sky fireball detection stations across Uruguay (Lat: -30° to -35°). The network was deployed between 2019 and March 2023, with stations separated by approximately 120 km to cover an area of ~180,000 km².

Hardware: The system utilizes locally assembled stations housed in waterproof enclosures. Initially, the network used Watec 902 H2 Ultimate analog CCTV cameras. In 2022, a major upgrade migrated the network to ZWO ASI 178MM digital cameras featuring a 6.4 Mpix CMOS sensor (IMX178) and a 2.5mm fisheye lens providing a ~170° field of view. Stations are connected to local nano PCs via USB 3.0 and utilize fiber-optic internet connectivity. A GPS receiver provides precise timing, and environmental monitoring (temperature/humidity) is managed via Arduino.
Software and Automation: The detection and processing pipeline is built using locally developed MATLAB and Python applications. The detection algorithm (bolidosGUI) relies on frame-to-frame image differencing. It triggers an event when the number of pixels exceeding a brightness threshold ( $n_{Diff}$ ) and the mean distance of these pixels to their centroid ( $r_m$ ) satisfy specific user-defined conditions, effectively filtering out large-area brightness changes like clouds.
Data Processing:
- Astrometry: A two-valued mapping function transforms pixel coordinates to horizontal coordinates (zenith angle and azimuth). Calibration uses a non-linear least squares approach (Levenberg-Marquardt algorithm) to determine plate constants, utilizing a synthetic star map from the Hipparcos catalogue.
- Photometry: The system performs aperture photometry on catalogued stars to derive atmospheric extinction coefficients and zero points. To handle bright, saturated fireballs, the methodology fits a 2D Gaussian to the sky-subtracted brightness to extrapolate flux beyond saturation levels. This allows for magnitude estimation using Gaia DR3 G-band magnitudes.
- Triangulation: Multi-station observations are processed using the Method of Planes (MOP) to reconstruct 3D trajectories and calculate geocentric radiant coordinates.
Citizen Science: The network is integrated into secondary education, with stations installed on school roofs. Teachers and students participate in station operation and manual video classification, which serves to train automatic classifiers.

Key Contributions and Results
The paper presents the first fully operational automated fireball network in Uruguay and provides a comparative analysis of the legacy Watec system versus the upgraded ASI system.

Astrometric Performance: The upgrade to the ASI camera significantly improved astrometric accuracy. The mean absolute error (MAE) for the ASI system was reduced to ~5 arcminutes (median residual 6.9 arcmin), compared to ~15 arcminutes (median residual 22.3 arcmin) for the Watec system. This improvement is attributed to the higher resolution (1.61 Mpix vs. ~0.4 Mpix) and the ability to extract a larger number of calibration datapoints from fewer video frames.
Photometric Validation: The photometric methodology was validated against known celestial bodies.
- Jupiter: The system estimated Jupiter's magnitude with a mean residual of 0.18 magnitudes relative to JPL Horizons ephemeris data.
- Full Moon: The system estimated the Full Moon's magnitude with a mean residual of 1.2 magnitudes. The paper notes a systematic difference where the computed G magnitude is fainter than the catalogue V magnitude, likely due to sky background over-subtraction and Gaussian fit bias on extremely saturated sources.
Fireball Detection Case Study: The pipeline successfully processed a bright fireball detected on October 29, 2022, across five stations. The event reached a peak absolute magnitude of -7.68, with a mean velocity of 32.1 km/s and a trajectory extending from 100.1 km to 53.5 km altitude. The trajectory was associated with the Southern Taurids (STS component), though the high end-altitude suggests it was not a meteorite-dropping event.
System Efficiency: The high sensitivity of the ASI camera allows the same short-duration video (1–3 seconds) used for detection to also serve for photometric reduction using field stars. In contrast, the Watec system often required generating longer, separate videos for photometry, a process that would create significant data storage challenges in a multi-station network.

Significance
The paper positions BOCOSUR as a significant contribution to the global infrastructure of automated fireball networks, specifically addressing the lack of coverage in South America. Its primary significance lies in demonstrating a cost-effective, replicable, and autonomous system that integrates high-resolution digital sensors with a citizen-science framework. The authors claim that the network is now fully operational (as of February 2023) and capable of providing precise pre-atmospheric trajectories and strewn-field calculations to facilitate meteorite recovery. Furthermore, the project serves as an educational tool, engaging secondary-level students and teachers in scientific research. The authors note that future work will focus on implementing automatic video classifiers, refining astrometric models for low-altitude events, and integrating atmospheric wind data to model the dark flight phase of meteoroids.

1. The "Sky Net"

2. The Upgrade: From "Old Glasses" to "Super-Vision"

3. How They Measure the Light

4. The First Big Catch

5. What's Next?

More like this