A Deep Dive into the Solar System’s Outer Frontier: Pan-STARRS Hunts for Distant Worlds

Holman et al. aim to answer one of astronomy’s lingering mysteries: are there still undiscovered planets in our solar system? The idea stems from the unusual orbits of certain Trans-Neptunian Objects (TNOs) like Sedna and 2012 VP113. These bodies move in ways that can’t be fully explained by the gravitational pull of known planets. Some scientists have proposed that these oddities might be caused by an unseen giant—possibly “Planet Nine.” This study contributes to the effort by searching Pan-STARRS1 telescope data for signs of such distant planetary objects.

The Pan-STARRS1 Telescope and Its Advantages

The Pan-STARRS1 telescope, located in Hawaii, has surveyed a large portion of the northern sky. Although its main mission is to detect near-Earth asteroids, its wide field of view and multi-year observational timeline make it a valuable resource for outer solar system searches. Unlike upcoming surveys like LSST, which will mostly cover the southern sky, Pan-STARRS1’s data gives researchers a complementary northern hemisphere view. The authors chose this dataset specifically because it extends across most of the ecliptic, the area of the sky where solar system objects are most likely to be found.

A Creative Approach: Synthetic Populations and Filtering Techniques

To overcome the challenge of identifying faint, slow-moving objects, the team created a “control population” of fake objects with realistic orbits and brightness. These were added to the existing Pan-STARRS1 data, allowing the researchers to test how well the survey could detect them. They accounted for issues like detector gaps, poor image quality, and background stars by applying strict filtering steps. By eliminating detections caused by noise, bright stars, or known stationary sources, they narrowed down the dataset to only the most promising candidates for new solar system discoveries.

Building Tracklets and Finding Orbits

The next step involved linking observations into “tracklets”—small groups of detections that represent the same moving object. Since distant objects move very slowly, the team had to gather data over many nights, sometimes years apart, to confirm a match. By carefully modeling the objects’ apparent positions and using a search algorithm that accounted for different distances and motions, they successfully identified objects whose paths aligned with gravitational models. They discarded any linkages with fewer than six tracklets to minimize false positives.

Results: What Did They Find?

In total, the team identified 692 real solar system objects, 109 of which were previously unlisted in the Minor Planet Center database. Impressively, they also detected 23 known dwarf planets like Eris and Gonggong, confirming the sensitivity of their method. While no new planets were found, the survey did demonstrate the ability to detect objects at distances up to 96 astronomical units (au), reinforcing that the Pan-STARRS1 dataset is capable of spotting very distant bodies under the right conditions.

Survey Simulator: A Tool for Future Discoveries

To better understand their detection capabilities, the researchers trained a machine learning model using their synthetic control population. This “survey simulator” can predict the likelihood that a particular object would be linked and discovered by their search. The model proved to be well-calibrated, meaning its predictions closely matched actual outcomes. This tool can now be used in future population studies, helping scientists understand what kinds of objects might still be hidden in the outer solar system.

Testing Planet Nine Hypotheses

Although Planet Nine was not discovered, the team used their simulator to test whether it could still be out there. They applied their detection model to a reference population of 100,000 possible Planet Nine orbits and found that most of them would have been detected by their search. The remaining plausible orbits are concentrated in the galactic plane—a difficult region to search due to the dense background of stars. This finding helps focus future searches on the most promising areas of the sky.

Conclusion: What Comes Next?

While the mystery of Planet Nine remains unsolved, Holman et al.'s study represents a major step forward in understanding the distant solar system. Their innovative use of synthetic populations and machine learning provides a robust framework for future searches. With upcoming surveys like LSST on the horizon, and continued refinements to their pipeline, researchers are well-positioned to either find new distant worlds or rule them out with confidence.

Source: Holman