"ExoClock Project IV: A Homogeneous Catalog of 620 Updated Exoplanet Ephemerides" (Kokori et al. 2026, ApJS, 283, 5) has been published in the Astrophysical Journal Supplement Series. I am a co-author, having contributed 30 transit photometry observations of several exoplanets to the ExoClock network. The publication is open-source and can be accessed at: https://iopscience.iop.org/article/10.3847/1538-4365/ae3238

When the European Space Agency's Ariel space telescope launches in 2029, its mission is to observe the atmospheres of 1,000 exoplanets by watching them transit in front of their host stars. To do so, it needs to know precisely when those transits will occur so that it can keep pointing from one star to the next at the right time.
 
It may seem like a straightforward scheduling problem, but it isn't. A planet's ephemeris, namely the information we have of the time it will take to transit and when its midtime will be, is determined from individual observations, and observations carry uncertainties. A period measurement that is off by even a fraction of a second compounds with every orbit. Over years, a small initial error can grow into an uncertainty of hours. By the time Ariel is ready to observe, a planet's predicted transit window could be so uncertain that valuable telescope time is wasted waiting, or the transit is missed entirely.
 
There is also the problem of transit frequency: some exoplanets transit every few days, others over weeks, months, or years, making observations harder to obtain and associated uncertainties larger. And where there are thousands of exoplanets, each transit running over several hours, it translates into thousands of hours of dedicated telescope time, and is not easy to find among competing priorities at professional observatories. 

The ExoClock initiative was launched in 2019 to address exactly this problem: too many transits to observe, and too few hours available at professional observatories. Its goal is to update the ephemerides for every Ariel candidate target before launch, and keep them updated through continuous monitoring through collaboration between a global network of citizen scientist astronomers whose scientific observations and inputs make the catalog possible. 

In practice this means determining which locations worldwide have the best observing windows for each transit, and relying on citizen astronomers to point their telescopes and observe them. The ExoClock network currently has 1,700 participants operating 1,600 telescopes. Each submitted light curve passes the same quality criteria:

This ensures that the data from different sources and telescope sizes can be accurately combined into a single homogeneous catalog. 

To date I have made 30 successful transit observations for Exoclock, on a  range of targets from well-studied hot Jupiters to recently confirmed TESS planets to planets flagged for transit timing variations. Several of these observations have been published in this paper, and others will be published in the subsequent release.
 
My workflow for each observation follows the ExoClock protocol: select a target from the observing queue based on scientific priority, schedule observations around the predicted transit window, image the star throughout ingress, transit, and egress, then reduce and analyse the resulting light curve before submitting it.
 
Some of the most interesting exoplanetary transits I observed:

WASP-148b was one of my highest-priority targets, rated ALERT by ExoClock, meaning its ephemeris uncertainty had grown to a level where upcoming transits were at significant risk of being missed. It is also a TTV-flagged planet in a multiplanetary system, where gravitational interactions with its companion WASP-148c are thought to drive the timing deviations. An ALERT rating means the observation window is time-sensitive: the longer it goes unobserved, the wider the uncertainty cone grows and the harder the planet becomes to schedule reliably for Ariel mission.

 TOI-2154b and TOI-3819b are TESS-discovered planets, the kind that present the most urgent monitoring challenge for ExoClock. TESS monitors each sky sector for only 27 days, compared to Kepler's four-year baseline, so TESS planets start with shorter observational baselines and larger ephemeris uncertainties that grow quickly without follow-up. Both targets are also on the dimmer end of the ExoClock list, requiring careful exposure management to maintain signal-to-noise across the full transit duration.

WASP-135b, HAT-P-7b, and TrES-3b are all flagged in the ExoClock catalog as showing transit timing variations (TTV). I observed all of them as these are the transit targets I find most useful, as they provide new data for my NEPTUNE project, which I discuss at the end. 

The co-authored paper, published in the Astrophysical Journal Supplement Series (ApJS, 283, 5, 2026), presents the fourth ExoClock data release integrating approximately 30,000 transit midtime measurements from ground-based telescopes of the ExoClock network, the space missions Kepler, K2, and TESS, and published literature. The result is an updated catalog of ephemerides for 620 exoplanets. 

The improvements are significant. On average, prediction uncertainties improved by roughly an order of magnitude — a median improvement factor of 7.9 times over the original ephemerides. Approximately 45% of planets required meaningful updates; without continued monitoring, the timing errors in those ephemerides would have grown to a level that would have disrupted Ariel's observing schedule and led to lost transit data. 

One of the technical challenges the paper addresses is the growing proportion of TESS-discovered planets in Ariel's target list. Currently 36% of the full ExoClock target list consists of TESS planets, and 61% of those require telescopes larger than 16 inches for reliable transit detection — pushing the limits of what small-telescope networks can efficiently cover. The paper introduces synchronous observations as a response: coordinating multiple small telescopes to observe the same transit simultaneously and combine their data. In the case of HD 191939b, combining nine telescopes smaller than 14 inches achieved a signal-to-noise ratio comparable to a TESS observation for a transit that would normally require a 25-inch instrument to detect. 

On the brighter side (pun intended), TESS planets tend to orbit brighter stars than Kepler candidates — Kepler targets extended to around magnitude 16 in the red, while many TESS targets are accessible at magnitude 12 or brighter — which does make their transits more accessible to citizen astronomers with smaller instruments.  

In single planet systems, the planets transit their host stars on a predictable schedule. But when another planet is nearby, its gravity tugs on the transiting planet, causing its transits to arrive slightly early or slightly late. Those small deviations from the expected schedule are transit timing variations, and they can reveal the presence of an unknown and unseen planet. Many exoplanets have been discovered this way and it is also a useful method to confirm existing exoplanets in multi-planetary systems.

For TTV analysis to be meaningful, the baseline ephemeris needs to be well-determined first; otherwise it is impossible to distinguish a genuine timing deviation from a poorly constrained period. Beyond correcting ephemeris drift, the dataset is now large enough to identify planets whose transit timing deviates from a simple linear pattern. The paper identifies 42 planets with statistically significant transit timing variations (TTV). Thirty belong to known multiplanetary systems, where gravitational interactions are the most natural explanation. The remaining twelve show long-term period changes consistent with quadratic ephemerides that may be possible signatures of orbital decay through tidal dissipation, or apsidal precession. They could also be signatures of  as-yet-undetected companions!

TTV analysis is directly relevant to my NEPTUNE research, which uses Transit Timing Variations to infer the presence of unseen exoplanets through the gravitational perturbations they leave on their neighbours' transit midtimes. The precision of that inference depends entirely on how well-determined the underlying ephemeris is, which is precisely what a catalog like this one provides. 

NEPTUNE began as a science fair project, winning a Gold Medal at the Canada-Wide Science Fair and the Third Grand Award in Physics and Astronomy at Regeneron ISEF 2025. It continues to expand with new observations, simulations, and analysis as datasets like the ExoClock IV catalog make the underlying ephemerides more reliable.   

All data products, tools, and updated ephemerides are publicly available through the Open Science Framework. The catalog is intended as a resource not just for Ariel, but for the broader community: any future transit observation, TTV study, or atmospheric characterisation campaign benefits from knowing when a planet's next transit will occur to within minutes rather than hours.
 
What makes ExoClock's model work is that rigorous quality standards make the contributions from a 12-inch backyard telescope combinable with data from professional observatories in Tenerife or Antarctica.  The network currently spans 1,700 observers worldwide including professionals, amateurs, and students working toward the same catalog, the same mission, the same deadline. 

The publication is open access: doi:10.3847/1538-4365/ae3238 

My ExoClock observations: exoclock.space/database/observations_by_observer#771