Physics of Space
Available: In-person: Princeton, NJ. Hybrid: everywhere.
A research lab for students who explore physics phenomena through data
For high school, middle school, and home school students
Mentor: Dr. Sergey Samsonau
Seize the Opportunity
In this lab, you do astronomy the way working astronomers do: from data. The telescopes are already built, and what they record is public. You pick a real object in the sky, measure one of its physical properties from the archives, and defend the measurement.
The Opportunity
Astronomy has an unusual problem: it measures faster than it can look. A single new survey telescope expects to record about 10 terabytes of sky images and issue millions of alerts every night. Whole-sky maps carry positions and brightness for almost two billion sources, and a digitized century of photographic plates holds tens of billions of measurements more.
A large share of modern astrophysics is done from such archives: a researcher queries one, writes analysis code, and publishes a measurement from data taken by an instrument they never touched. The questions are physics: what is this object, how does it move, why does it change.
The archives also run deeper than anyone has looked. Of the roughly 341,000 entries in the asteroid light-curve database, only about 31,000 carry a rotation period reliable enough for statistical work.
What Students Do
Finding your question is the first piece of the work. The mentor helps you pick a corner of the field you care about, read into its literature, and shape a question that is measurable and yours.
From there the work is computational. Most projects run in Python and the standard tools of the field (Astropy, Lightkurve, TOPCAT, Aladin). Finding a period is Fourier analysis, but the point is rotational dynamics; counting craters is statistics, but the point is the age of a surface. The daily craft is querying, fitting, quantifying uncertainty, and deciding whether a signal is real or an artifact.
Then you write it up and aim to publish. Genuine results can be submitted to the SoTS journal and presented at the SoTS conference.
Recent Work in This Field
This field is active, and much of its recent literature is free to read. A few studies from the areas this lab works in:
The spin of asteroids
- Gaia's third data release carries photometry for over 150,000 asteroids; from it, researchers reconstructed unique spin states and low-resolution shape models for about 8,600 (Ďurech and Hanuš, 2023).
- A 2024 study derived asteroid rotation periods by combining a ground survey with a space telescope, and quantified how often survey-derived periods come out unreliable (Gowanlock and colleagues, 2024).
- The Minor Planet Bulletin publishes new rotation period measurements. One recent paper reported light curves of 22 asteroids (Stone, 2025). Another in the same issue, with a high school student as first author, measured the period of asteroid 4222 Nancita with a 13-centimeter telescope (Chen and colleagues, 2025).
A century of starlight
- One astronomer measured the star KIC 8462852 (“Tabby's star”) fading by about 0.16 magnitudes per century across the Harvard plates (Schaefer, 2016); another team countered that at that level the plates themselves can drift (Hippke and colleagues, 2016).
- A search of a single patch of sky in the plate archive found three giant stars swinging by a full magnitude over decades. They resemble no reported class of variable star, and no model explains them (Tang and colleagues, 2010).
- The completed DASCH digitization was released in 2024, with a documented Python package for pulling any star's light curve (Williams, 2025).
Star clusters in Gaia
- An all-sky census built from Gaia's third data release recovered 7,167 star clusters, 2,387 of them candidate new objects (Hunt and Reffert, 2023).
- Another study showed that dozens of “clusters” that sat in catalogs for decades are chance alignments of unrelated stars (Cantat-Gaudin and Anders, 2020).
- Gaia astrometry revealed 800-parsec-long tidal tails around the Hyades (Jerabkova and colleagues, 2021), and tails around Praesepe as well (Röser and Schilbach, 2019).
Surfaces that still move
- When experts counted craters on the same lunar images, their totals varied by up to about 45 percent (Robbins and colleagues, 2014), a spread that matters because crater counts are how planetary surfaces are dated.
- A machine-learning sweep of Lunar Reconnaissance Orbiter imagery mapped 136,610 rockfalls across the Moon (Bickel and colleagues, 2020).
- Repeat orbital image pairs showed megaripples migrating at two Martian sites (Silvestro and colleagues, 2020).
- Dust devil motion, measured across orbital images, revealed near-surface winds on Mars stronger than atmospheric models predicted (Bickel and colleagues, 2025).
Objects in Earth orbit
- About 40,000 objects were being tracked in orbit in 2025, roughly 11,000 of them active satellites, alongside an estimated 128 million fragments too small to track. In low orbit they meet at about 10 kilometers per second, fast enough that a centimeter-wide chip is dangerous (Space Environment statistics, 2025).
- Two events account for much of the debris: the 2007 destruction of the 880-kilogram Fengyun-1C satellite in a weapons test left about 1,800 cataloged fragments (Pardini and Anselmo, 2007), and the 2009 Iridium-Cosmos event, the first collision of two intact satellites, left more than 2,000 (Wang, 2010).
- A satellite's orbit decays through atmospheric drag, and the drag rises when solar activity heats the upper atmosphere. Tracking 523 Starlink reentries from 2020 to 2024 in the public orbital catalog, one study found they fell faster during geomagnetic storms (Oliveira and colleagues, 2025).
- Reentering spacecraft vaporize high in the atmosphere. About 10 percent of stratospheric aerosol particles now carry metals from them, and for aluminum the human-made input already rivals the natural dust that meteors supply (Murphy and colleagues, 2023; Schulz and Glassmeier, 2021).
- Bright satellites also cross astronomers' images. Photometry of more than 100,000 Starlink observations measured them near naked-eye brightness, and a darkening treatment cut that only partway (Mallama, 2021; Fankhauser and colleagues, 2023).
The sky from the ground
- About 83 percent of the world's population lives under light-polluted skies, and the Milky Way is hidden from more than a third of humanity (Falchi and colleagues, 2016).
- It is changing fast. Naked-eye reports from 51,351 citizen observers put the sky brightening at about 9.6 percent a year from 2011 to 2022, enough to take a location from 250 visible stars to about 100 across a childhood (Kyba and colleagues, 2023).
- Satellites see the ground brightening at only about 2.2 percent a year, and reconciling that with the naked-eye rate is unsettled: the gap traces to the shift toward blue LED light and to sources satellites cannot see (Bará and Castro-Torres, 2025).
- Amateurs supply the data. Globe at Night collects naked-eye star counts worldwide, and the Global Meteor Network has recorded more than 220,000 meteor orbits from over 450 low-cost cameras across 30 countries (Vida and colleagues, 2021).
The Archives
The data behind all of this is public. Among the archives this lab works from:
- Gaia Archive (ESA): positions and brightness for 1.8 billion sources, with proper motions for 1.46 billion.
- DASCH (Harvard): a digitized century of photographic plates, the 1880s to 1992.
- MAST (NASA): space telescope data, including the Hubble and JWST archives and the TESS and Kepler light curves.
- ALCDEF and the LCDB: asteroid light curves, and the database of rotation periods with quality ratings.
- ZTF: public data releases from a repeating survey of the northern sky.
- LROC and HiRISE: high-resolution imagery of the Moon and Mars, including repeat coverage of the same ground.
- CelesTrak: free orbital element sets and the satellite catalog for tracked objects in Earth orbit.
- Globe at Night and the Global Meteor Network:citizen naked-eye sky-brightness reports, and meteor orbits from amateur cameras.
- NASA Physical Sciences Informatics: data from physics experiments run on the ISS, open for reanalysis.
Mentor
- Physicist: PhD in Physics (experiment and simulation), MS in Theoretical Physics, 25 years in the field
- Trained 100+ students in research methodology at NYU
- Built and directed research labs program at PRISMS (top USA high school)
- Led AI projects at financial, medical, and educational institutions
- Conceptualized and organized AI Meets Science NYC-metro area conference
- Developer of original research education methodologies
Prerequisites
- You like describing and exploring phenomena through data, and statistics makes sense to you
- Experience in a computational language (Python, R, MATLAB, Wolfram, or similar) is welcome but not required
- Curious about why objects in the sky behave the way they do, not just how they look
- Willing to read into a real literature and design your own study, not follow pre-made assignments
- Commitment to weekly meetings and work between sessions
No prior research experience, astronomy background, or telescope is required.
General Details on SoTS Research Labs
Enrollment
Available in In-person and Hybrid formats, in Fall, Spring, and Summer semesters. See Princeton Labs enrollment for full details and the schedule call.
The data has already been taken. The question is still yours.
Outside the Princeton, NJ area? Hybrid lets students anywhere participate in authentic weekly research: Zoom sessions plus a few half-day in-person visits per semester. See Princeton Labs enrollment for details.