Ammonia World statistics in a few pictures

Heavy Johnson · Jul 6, 2019

So I got bored and crunched numbers on Ammonia worlds based on the data helpfully provided by EDastro.com. Here's what you get:

Your best chances for finding AWs are in K, M, and F class systems. G and A less so, and everything else is not even worth putting on the graph.

This really just confirms the above data. AWs are mainly in d and c mass systems, with some stragglers in b mass systems. Those are also the mass types you expect to find K-F type stars. If in the future we can filter the route plotter by system mass, this might be helpful.

If you're lucky enough to find an AW moon around a gas giant, it's most likely from a Class III gas giant. Amusingly GG with ammonia life are less likely to harbor AW moons than GG with water based life. Note: There are ZERO known AW around a Class V gas giant. First explorer to find one gets bragging rights.

Ammonia worlds radius is strongly focused in the 4000-5000 km range, and then sharply drop off.

Any other statistics on AWs anyone is interested in?

Rooster Podunk · Jul 6, 2019

"There are ZERO known AW around a Class V gas giant. First explorer to find one gets bragging rights."

digs out the bucket list...yes right here on page 598...got it.

Sapyx · Jul 7, 2019

I think "Class V giant" and "ammonia world" are in mutually exclusive formation zones - that is, Class Vs only form in regions that are too hot for ammonia worlds. Just like you never find an ammonia world moon orbiting an Earth-like, or even a TFC. Or vice-versa.

The sixty-odd AWs found around Water Giants I find curious, given how relatively few WGs there are.

Leif_Erikson · Jul 7, 2019

Sapyx is right- class V giants will be too close to the star for any Ammonia worlds nearby - not sure I've seen any moons or worlds orbiting a class V ?

As of last year I know there are only two Ammonia Worlds orbiting a White Dwarf- I found one and visited the other one.

Heavy Johnson · Jul 7, 2019

There are currently 41 AW around all the different types of White Dwarves. Interestingly many of these are in the high mass codes (d-h).

Heavy Johnson · Jul 7, 2019

Wrupeae AA-A h220 is an interesting example: An AW in a very close orbit (0.6 d orbital period) of a DC-type white dwarf; the system also has a neutron star; and the primary is a black hole!

Leif_Erikson · Jul 7, 2019

Sorry, should have been more specific- only two Ammonia Worlds reported on EDSM orbiting the main White Dwarf in a system.

Heavy Johnson · Jul 7, 2019

I show 8 systems with an AW orbiting the main WD:

Bleia Eohn SZ-G d10-0

Thailoi PM-M d7-5

Rhuedgo KW-M d7-729

Eulail MH-V d2-38

Flyeia Eock LS-R d5-1

Dryi Broae OZ-D d13-2066

Aemonz NI-T d3-1976

Plooe Aesms IT-Z d13-2

Leif_Erikson · Jul 7, 2019

Heavy Johnson said:
I show 8 systems with an AW orbiting the main WD:

Bleia Eohn SZ-G d10-0
Thailoi PM-M d7-5
Rhuedgo KW-M d7-729
Eulail MH-V d2-38
Flyeia Eock LS-R d5-1
Dryi Broae OZ-D d13-2066
Aemonz NI-T d3-1976
Plooe Aesms IT-Z d13-2

Found Rhuedgo and visited Dryi Broae. Now I have another 6 to visit!

schlowi123 · Jul 7, 2019

Interesting analysis.

You wrote:

Heavy Johnson said:
Your best chances for finding AWs are in K, M, and F class systems.

Do you mean the main star of a system? If yes, you show a graph in which it seems that the < parent star > is used on the y-axis. These need not to be the same.

I did an analysis in which I use the main star of the system as criterion and I get mostly the same numbers. With two exceptions in which the difference is not very large, but noticable:

It doesn't make much sense to look for the main star of a system due to the same reason I mention above. But this is the only thing that can be filtered for in the star map.

Anyway, I stumbled over that and what I actually wanted to say is the following: you need to divide by the number of stars that have been found of a given (main) star type to state the chances to find an AW in such a system.

This however changes the results significantly:

Edit: The ordinate is of course in percent!

So with the star map filter it actually is better for you to go for G and A star systems than for M stars.

Anyway, these numbers are over the WHOLE 2018-06-18 EDSM dataset. This contains a lot of old data in which many many main stars got not scanned and are thus unknown. I will do the analysis again, tomorrow with data just after the advent of the fss.

I also may have made a gross mistake and all I say may be totally wrong.

Edit: AAAGRHAGRHG … typo in the y-axis label

Heavy Johnson · Jul 7, 2019

But this is the only thing that can be filtered for in the star map.

Indeed, and multiple star systems are common. But that's all we can filter on.

you need to divide by the number of stars that have been found of a given (main) star type to state the chances to find an AW in such a system.

That's a reasonable point, and thank you for pointing it out.

marx · Jul 8, 2019

About code H, and the moons. The temperatures of atmospheric moons are bugged, and less than they should be. This does help with the formation of the "sensitive" bodies there, especially ELWs, but AWs too. For example, out of the 151 ELWs in mass code H on EDSM (on 2019.05.30), 141 were ELMs. I haven't looked at AWs and AMs for a long time, but back at the end of 2017 (see here), the AM / total ratio was 37 / 57.
Since you have the data processed already, could you check how that looks now?

Heavy Johnson · Jul 8, 2019

marx said:
About code H, and the moons. The temperatures of atmospheric moons are bugged, and less than they should be. This does help with the formation of the "sensitive" bodies there, especially ELWs, but AWs too. For example, out of the 151 ELWs in mass code H on EDSM (on 2019.05.30), 141 were ELMs. I haven't looked at AWs and AMs for a long time, but back at the end of 2017 (see here), the AM / total ratio was 37 / 57.
Since you have the data processed already, could you check how that looks now?

I'm not sure the data dump includes the necessary information to determine AW vs AM, but I'll look after I'm done with classes for the day.

marx · Jul 8, 2019

Heavy Johnson said:
I'm not sure the data dump includes the necessary information to determine AW vs AM, but I'll look after I'm done with classes for the day.

The planet ID does. If it ends with a letter, it's a moon.

Heavy Johnson · Jul 8, 2019

What were you looking exactly for, marx? The ratio of AM to AW, for all ammonia-type planets? Or just in class H?

foobaron76 · Jul 8, 2019

I would be interested in the distribution of the mass of AWs, to see if it correlates nicely with the radius distribution.

marx · Jul 9, 2019

Yep, in mass code H only. Although it never hurts to have more data

Back earlier, the differences between codes C and D were rather small (39% vs 43%) for AWs, but so was the sample size compared to today's. (23,536 AWs in total) For AMs, however, 81% were in code D.

schlowi123 · Jul 9, 2019

As promised, the data for before and after the introduction of the FSS.
The absolute numbers change of course, but the change is (as one could have suspected) more like the subtraction of a baseline:

However, the chances to find an AW with a given main star type don't change a lot (in the all-data vs. after-FSS-data case [red and blue bars]):

The latter is rather interesting. I would have expected that the changes are more pronounced. On the other hand, the data of the blue bars is included in the red bars and we are talking about huge numbers. So any significant changes probably . Maybe most people that scanned AW's before the FSS also scanned the main star. Just because I didn't do that doesn't mean the majority wasn't more thorough.

Hence, I investigated how it looked like before the FSS appeared (green bars) and behold, the probabilities change more in some cases.
Most significant is the case with a Neutron star as the main star. I assume that most people that go to them want to continue fast. Thus, before the FSS even if they saw an ELW, they did not scan it. But with the FSS it's taking much more less time, thus they started to scan them.
That may also be the case for the A and F cases. Is it possible that the goldilocks-zone is further away for them than for the other cases? In this case some explorers may not have bothered supercruising there while with the FSS these are included, too.

schlowi123 · Jul 9, 2019

Ah yes! The chance to find an AW in a F system is double as high as stumbling over one without any filtering at all (the latter is approx. 0.82 %)

schlowi123 · Jul 9, 2019

So the mass code distribution was of interest. Here it is per star type:

So the vast majority of AW's (in absolute numbers) can be found in certain mass code systems. I would say, that doesn't come as a surprise, since certain stars are prevalent in certain mass code systems.

However, the probability distribution looks (very) interesting:

For mass codes / star type combinations without a bar, it means that the number of AW's in such systems was zero. The discussion is of course meaningful just for the cases with actual data.

For K type (main) stars it does almost not matter what the mass code of the system is.
Dito for Neutron and F stars.

However, if one wants to find an AW around an M type star, the odds are higher in mass code < e > systems. Although … there are just 6 of them … so this is maybe an artifact.
This is probably also true for the higher probably of the G-type / e-masscode combination (9 AW's in this category).
For AW's around A type stars however one should search in mass code < d > systems (probably NO artifact since the number of AW's found in A-type / e-masscode systems is > 100).

So, if one want's to find Ammonia worlds, one should look in systems with an F-type main star and masscode < d >.

Edit: this analysis was performed JUST with data AFTER the introduction of the FSS!