Full HARPS 2017 dataset now available

We have been a bit busy organizing the data, but the final HARPS dataset from Red Dots is now available for downloads at


If you want make use of the data for scientific publications, remember that this is still preliminary, and that we welcome contributions and open discussions on this dataset. We mentioned the possibility of a second strong signal in Proxima’s RVs. Instead of forcing our conclusions into you, we let you take a look at all the observations and try to draw your initial conclusions (if you are new, we suggest using the systemic tool) For those with technical expertise in high resolution spectroscopy and science formats, you can access the ESO reduced files in the Proxima/Spectra folder, and the corresponding up-to-date radial velocities in Proxima/timeseries/.

Periodogram search on the ‘Red Dots 2017’ data set only. As in last year, the signal at ~11 days is significant in the new set, adding further robustness to the detection of Proxima b. This also implies we are likely to reach similar sensitivities on the other two stars (Barnard’s and Ross 154). Exciting. Combination with previous data might reveal additional signals, but these analyses will take more time and thinking. Image credits : Guillem Anglada/Red Dots

A very quick analyses of the 2017 only data, shows that the signal of Proxima b is again detected with the new data set only! This was not the case for a while (data from first weeks was not as good as last year), but it looks like that in the end Proxima b’s signal remains in healthy confirmed state.

The spectrum taken on Sep 24th and a few other nights was not usable due to contamination by Sun-light. This typically happens due to twilight observation, moon proximity and/or the presence of high clouds scattering light. Red is the spectra registered on Sep 24th, and black is data registered in a good night with dark conditions. Image credits : Guillem Anglada-Escude/Red Dots

It is worth mentioning that we dropped three spectra because they were badly contaminated with solar-like stray-light. That happens because they were taken a bit too early after the sunset, because the moon was near the star on the sky, presence of high clouds scattering sun & moonlight, or a combination of the three. These spectra are in Proxima/Spectra/contaminated/ in case anyone has a use for them.

The photometric datasets are almost ready, but we need to clean them up and organize a little. Data is coming from more than a dozen observatories and the formats are not fully consistent with each other. Lot’s of spreadsheet work to do!

Be the first to comment

Leave a Reply

Your email address will not be published.