So, thanks to the RFI Team, we now have a confirmation of something I suspected before (see here or here) - that the lack of new additions is the result of no new discoveries and lack of new data about past discoveries since March. What is more, we now know that "new exoplanets are discovered and analyzed for potential habitability every week." (see latest clarification to this question).
I looked at the database of updates to the catalog, which I created using Archive.org (Google sheet, my comment). We can see that since 2016-12-05 we had about 40 additions, and since 2020-01-16 we had about 15 additions. Clarification to the resolution criteria makes it clear that only additions to the catalog count even if some exoplanets were removed. Since my database shows that in the past some exoplanets were also removed from the catalog it is possible that a number of additions is, in fact, a bit higher in that period since there were not that many updates per year and if before 2024 in any update there were both additions and removals of exoplanets from the catalog the total number of exoplanets in the archive would not show this - you can see that removals happened in the past (data about exoplanets is being updated with new observations and new calculations - this can lead to removal or addition of previously discovered exoplanet like with K2-3 d which was added to the catalog this year but was discovered in 2015) .
I used Poisson distribution to calculate the probability of 4 or more exoplanets being added using these two rates of additions from my database mentioned above at least 40 since the end of 2016 and at least 15 since January 2020. That gives us approximately 6.30% (if we use the end of 2016 rate) or approximately 1.35% if we use the rate since January 2020. We should also have in mind that most of the discoveries are announced in scientific articles (in case of some teams they may be added to the databases like NASA archive at the same time). I previously tried to establish if there is any irregularity of publications of these articles, and academic holidays, and especially August may be a time when fewer articles are submitted to the scientific papers in general. Another such period may be December (for example see this data). At the same time I also checked submission dates of preprints of articles about exoplanets discovered in 2023 which were added to the catalog this year, and it is not like most of them were from January (assuming I found the right papers for each of them) - however 2 are from January: TOI-700 e - Submitted 9 January, 2023; originally announced January 2023 (url) TOI-715 b - Submitted 10 May, 2023; originally announced May 2023. (url) HN Lib b - Submitted 31 May, 2023; originally announced May 2023. (url) Wolf 1069 b - Submitted 2 February, 2023; v1 submitted 6 January, 2023; originally announced January 2023. (url) TOI-904 c - Submitted 23 October, 2023; originally announced October 2023. (url) GJ 367 d - Received 2023 February 16; revised 2023 June 15; accepted 2023 June 21; published 2023 September 14 (url)
This is a very small sample (we could check the articles for all of the 70 planets in the catalog if someone has more time (I may do that in the next forecast), but as of now I only wanted to confirm that it is not like all are published in January or in Q4. Looking at publication dates of preprints about other articles may be a good idea, but I already spent 6 hours on this question today, and I also worked on it in the last week of so before the clarification came. I created a python script for getting data from NASA catalog and checking for new candidates - I am still not satisfied with my calculations being confirmed as right, but this time it got 4 of 6 right added this year and discovered in 2023 right - unfortunately NASA exoplanets archive is the main but not the only source of information they use, so my data is not the same as the data used by HWC).
It's not so simple, and our data sample is almost by definition small in order to draw such conclusions. Based on the data compiled for a previous comment of mine (largely based on previous excellent detective work by @michal_dubrawski):
Only 2 planets were added to the catalog in 2023, both of them in January.
So, it is (very...) easy to imagine someone in Nov-Dec 2023 arguing that "obviously,the pace of new discoveries is slowing"
... and then, we have 6 (!) new additions in January 2024 alone...
So, thanks to the RFI Team, we now have a confirmation of something I suspected before (see here or here) - that the lack of new additions is the result of no new discoveries and lack of new data about past discoveries since March. What is more, we now know that "new exoplanets are discovered and analyzed for potential habitability every week." (see latest clarification to this question).
I looked at the database of updates to the catalog, which I created using Archive.org (Google sheet, my comment). We can see that since 2016-12-05 we had about 40 additions, and since 2020-01-16 we had about 15 additions. Clarification to the resolution criteria makes it clear that only additions to the catalog count even if some exoplanets were removed. Since my database shows that in the past some exoplanets were also removed from the catalog it is possible that a number of additions is, in fact, a bit higher in that period since there were not that many updates per year and if before 2024 in any update there were both additions and removals of exoplanets from the catalog the total number of exoplanets in the archive would not show this - you can see that removals happened in the past (data about exoplanets is being updated with new observations and new calculations - this can lead to removal or addition of previously discovered exoplanet like with K2-3 d which was added to the catalog this year but was discovered in 2015) .
I used Poisson distribution to calculate the probability of 4 or more exoplanets being added using these two rates of additions from my database mentioned above at least 40 since the end of 2016 and at least 15 since January 2020. That gives us approximately 6.30% (if we use the end of 2016 rate) or approximately 1.35% if we use the rate since January 2020.
We should also have in mind that most of the discoveries are announced in scientific articles (in case of some teams they may be added to the databases like NASA archive at the same time). I previously tried to establish if there is any irregularity of publications of these articles, and academic holidays, and especially August may be a time when fewer articles are submitted to the scientific papers in general. Another such period may be December (for example see this data). At the same time I also checked submission dates of preprints of articles about exoplanets discovered in 2023 which were added to the catalog this year, and it is not like most of them were from January (assuming I found the right papers for each of them) - however 2 are from January:
TOI-700 e - Submitted 9 January, 2023; originally announced January 2023 (url)
TOI-715 b - Submitted 10 May, 2023; originally announced May 2023. (url)
HN Lib b - Submitted 31 May, 2023; originally announced May 2023. (url)
Wolf 1069 b - Submitted 2 February, 2023; v1 submitted 6 January, 2023; originally announced January 2023. (url)
TOI-904 c - Submitted 23 October, 2023; originally announced October 2023. (url)
GJ 367 d - Received 2023 February 16; revised 2023 June 15; accepted 2023 June 21; published 2023 September 14 (url)
This is a very small sample (we could check the articles for all of the 70 planets in the catalog if someone has more time (I may do that in the next forecast), but as of now I only wanted to confirm that it is not like all are published in January or in Q4.
Looking at publication dates of preprints about other articles may be a good idea, but I already spent 6 hours on this question today, and I also worked on it in the last week of so before the clarification came. I created a python script for getting data from NASA catalog and checking for new candidates - I am still not satisfied with my calculations being confirmed as right, but this time it got 4 of 6 right added this year and discovered in 2023 right - unfortunately NASA exoplanets archive is the main but not the only source of information they use, so my data is not the same as the data used by HWC).
Comment deleted on Oct 14, 2024 02:55PM UTC
@Plataea479
It's not so simple, and our data sample is almost by definition small in order to draw such conclusions. Based on the data compiled for a previous comment of mine (largely based on previous excellent detective work by @michal_dubrawski):