Hi!
I’m planning my observations for the next few months. I’ve always used the Transit Scheduler from the ExoClock campaign, but now I decided to try the Transit Finder from Swarthmore.
There is one specific planet I’m interested in, WASP-19b. But there is something that I don’t understand: for the same day each tool shows different observing times.
In the Transit Scheduler the observation starts at 2026/01/06 - 00:43 and ends at 2026/01/06 - 04:43.
In the Transit Finder the observation starts and 2026/01/06 - 04:43 and ends at 2026/01/06 - 08:10.
In both tools I used the same latitude and longitude (-34.605194, -58.4346), set the observatory timezone to UTC, set 1 hour before ingress and after egress, and elevation to 20°.
For this planet, I understand the Transit Scheduler shows the ephemeris from Kokori et al. 2025, while the Transit Finder shows the ephemeris from Sodickson & Grunblatt 2025.
Any idea why such a big difference? Who is right?
Thanks,
Joaquín
Hi Joaquín!
Take this with a grain of salt, because I’m no expert, but it seems that Sodickson & Grunblatt (2025) are searching for orbital decays, as you can see on this O-C graph in their paper:
They say that WASP-19 b, being a hot Jupiter is subject to tidal orbital decay and found that the best model −at least for them− says that it’s period is decaying at a rate of −3.90 ± 0.37 ms yr⁻¹. This is a very little number, so I don’t think that this suffice for explaining the difference.
Now… Kokori et al. (2025), if I read correctly are saying that WASP-19 b (and others exos) requires period corrections, but they are using observations taken up to 2023. Since both are using models (obviously), both could be wrong… but, since Sodickson & Grunblatt are using new observations, I tend to trust more on them.
But… experience and observation always have the last word.
By the way, I tried to reproduce your Swarthmore output and WASP-19 b is conspicuous by his absence. Now I’m very confused!
Fortunately, you can use Exoplanet archive to produce ephemerids, selecting your favorite literature:
https://exoplanetarchive.ipac.caltech.edu/overview/WASP-19%20b#planet_WASP-19-b_collapsible
Best regards and happy 2026,
Carlos
1 Like
What I see is that in the Transit Finder capture, local time and UTC are the same. Is that really the case at your observing site? That might explain the difference, since the time of the beginning of the observations is suspiciously different by exactly 4 hours. Is your time zone UTC-4?
I took the chance to update WASP-19 elements in VSX based on the most recent paper (2025AJ…170..133B).
The resulting VSX ephemeris indicates that the UTC times are 01.48 - 03.24 (thus starting 00.48 if you want that to be 1 hour earlier).
So then it is not a UTC-Local time issue but I do think something is wrong with the time display on the Transit Finder results.
There are Transit Timing Variations present in this system but that is not related to the 4 h difference.
Cheers,
Sebastian
1 Like
Carlos,
Thanks for the detailed explanation and for the link to the Exoplanet Archive to produce ephemerides. It’s a really interesting tool!
Until now, I’ve always used the predictions from the ExoClock campaign, and they’ve always worked. For now, I’m going to keep using them, even if they are based on older data.
Thanks,
Joaquín
Sebastián,
Thanks for the input and for updating the ephemeris on VSX.
In the Transit Finder, in the observatory section, I had to manually enter my coordinates, and I also had to select the UTC timezone because the drop-down list doesn’t include mine, which is UTC-3.
I don’t think there is a problem with the time display in the Transit Finder results, because I only have this issue with WASP-19b. For example, the ephemeris for WASP-43b is the same in both tools, even when I use UTC. The difference with this planet is that both tools use the ephemeris from Kokori et al.
The new ephemeris on VSX is almost identical to the one in the Transit Scheduler, so I’m going to use that.
Thanks for your help,
Joaquín
OK, I found the explanation.
The main table in Sodickson & Grunblatt 2025 shows periods rounded to three decimal places (0.789 d), so even when the epoch T0 given to three decimal places might be enough, you can’t use such period to create an ephemeris. Especially when the epoch is almost 13 years ago (April 20, 2013, 5886 transits have elapsed until January 6, 2026).
They give the unrounded period only in their Fig. 8 (0.78883901 d). If you use this figure, “your mid-transit” happens exactly at the same time as the VSX ephemeris predicts: 02.36 UTC.
When we use 0.789, it happens at 06.24 UTC (it is 06.22 in your capture, but that’s negligible, maybe some rounding in the routine creating the ephemeris).
Moral: a rounded figure can never be used to create an ephemeris.
Maybe they just took the data from the table in an automated way.
Cheers,
Sebastian
Great! Thanks for looking into it. I’ve also been able to get the same values as you — I’ve learned a lot today.
