Analysing the ELW data from EDDB/EDSM

[marx]

Yep, that's pretty well established, and that picture of yours sums it up excellently. The reason is likely to be how the planets are formed in various types of systems - Earth-likes orbiting cooler stars close, Earth-likes orbiting hotter ones farther out, and so on. Then there are the Earth-like moons too, of course, but thankfully, they aren't that numerous.
It would be nice to look at ELWs broken up into categories by the stars they are orbiting, like on the ELW list, and then analyze those separately, but the EDDB data doesn't include that easily. Surprisingly often, people don't even scan the arrival star in the system.

Did you happen to look at ammonia worlds too, by the way? They show this even more pronounced, although given that they have much larger possible differences in their characteristics (going as far as to not have an atmosphere), that's little wonder there.

 

[MattG]

Matt would it be possible to do a map of just ringed ELW's?

Not wanting to hijack this thread, but I'll have a look. Hard to make it relevant though as I strongly suspect it'd just look like the normal ELW map but less sparse. Probably makes more sense to plot something like ringed ELWs as a percentage of all ELWs in a given area...

 

[MattG]

It shows three clear "sets" of ELWs:
- the low-gravity, low-pressure set, below 0.87 G.
- the high-gravity, low-pressure set, above 1.03 G and around 0.75 atms pressure
- the medium-gravity, high-pressure set, 0.91 to 1.17 Earth masses and with pressures rising steeply from 1 to 4.25 atms.

Notice the big void in the graph around the 1.00-1.00 point and the general sparsity of anything in the 0.90-1.10 "box" around that point. Most of the ELWs around 1 Earth-gravity have much thicker atmospheres (2.5 to 4 atms).

That's a fascinating graph, great job.

 

[Sapyx]

At first, I thought these clear, distinct categories of ELWs might be a remnant of the old FE2/FFE sub-classification of ELWs as different types (normal, Ice World, Jungle World and Desert World) so I tried subdividing the graph above by temperature into four ranges: subfreezing (260-273K), cold (273-288K), warm (288-302K) and hot (303-320K). I saw no significant changes (all temperature ranges showed all three major lobes) so that scuttled that theory, but I did notice a couple of curious things on these graphs:
- All of the scattering of high-pressure, high-gravity worlds, including the thin line above the main Set 2 line, are sub-freezing. I would put this down to the "argon effect": argon only occurs on cold worlds, and would increase the atmospheric pressure above "normal" without changing the overall habitability.
- Almost all of the Set 2 worlds above 1.6 gravities are in the "Hot" category.

 

[Jackie Silver]

I'm going to hazard a guess that the distinction in ELW types comes from their material Ice/Rock/Metal split, and from their status as moons or free-orbiting planets? Moons always seem to have a lower metal percentage than their parent bodies.

Or possibly, due to the presence of trace greenhouse gasses in the atmospheric mix?

 

[marx]

I'm going to hazard a guess that the distinction in ELW types comes from their material Ice/Rock/Metal split, and from their status as moons or free-orbiting planets? Moons always seem to have a lower metal percentage than their parent bodies.

Or possibly, due to the presence of trace greenhouse gasses in the atmospheric mix?

Hm, probably not the ice material split though. Ice only appears in an ELW composition if there's argon present in the atmosphere, and only if there's a high enough partial pressure. (So argon in atmosphere -> might be ice, ice in composition -> there's definitely argon) The metal and rock ratio is a good idea though, gonna have to look into these later... But then, the question is, where does all the ice come from? Especially with those rare Earth-likes which are mostly ice.

 

[Sapyx]

About the characteristics of Earth-likes:
...
Also, speaking of the Forge hating things, note how rare eccentric orbits are for ELWs. Little wonder though....

This is one thing I find interesting: not that most ELWs have minimal eccentricity, but that a tiny few of them have seriously out of whack eccentricities, yet are still ELWs. A highly eccentric moon or co-orbital planet I could understand. And I suppose a highly eccentric orbit around a neutron star could remain inside a neutron star's very wide goldilocks zone. But a world around a sole main-sequence star? That's worth checking out. There's one only a few thousand LYs from Colonia, in a kind-of-homewards direction. I guess I've found my next exploration waypoint! See if I can find out if these things are really inhabitable comets. [big grin]

Here's my list of "top five" highly eccentrics: the first four are sole-stars (two neutrons, and two main-sequence). The bottom one is in a multi-star system but is the record highest eccentricity.

Zunoa UO-Z d13-113	5	0.5075
Dryau Aowsy EX-J d9-5293	4	0.5482
Eephaik HL-X d1-16	4	0.8322
Pheia Briae CE-B d1852	1	0.8995
Byoomao ES-I d10-9066	A 1	0.9271

 

[_RUKUS_]

Greetings CMDRs,

I began a little project to try to determine which areas along the galactic planes ELWs were most likely to form, and decided to post all my findings here per suggestion. Apologies for cross-posting if you've already seen it elsewhere.
Basically the idea was to take all the data I could find and establish intervals above/below the y coordinate.

The first data set I found was CMDR Marx's list of ELWs, containing coordinates for approximately 6,600 ELWs. This was the initial distribution and percentage found within each interval.

After coming up with a figure of from this dataset, it seemed likely that the findings were skewed due to explorers influencing intervals +/- 1,500 Ly as well as the fewer number of total systems explored in said intervals.

Taking all of EDDB's 15 million and some odd systems, I broke down how many had been discovered in each interval for reference.

After finding a way to attach coordinates to a larger data set of Earth-likes, this was the result. The percentages represent the number of systems contain ELWs in each interval, divided by the total systems discovered in each respectively. This was compiled using ~32,000 systems containing ELWs, so while some of the intervals with fewer total systems discovered may skewed it overcomes some of the other issues presented with smaller data sets.

My next idea (beyond getting data for ALL ~43,000 ELWs) is to put these data points into a 3D map so that we can take a closer look at patterns. Also, I think it'd be really cool.

o7

 

[Redfox]

My next idea (beyond getting data for ALL ~43,000 ELWs) is to put these data points into a 3D map so that we can take a closer look at patterns.

Going 3D sounds interesting, looking fwd to that. Here are a few 2D-views (incl. 26k ELW) from a while back: https://forums.frontier.co.uk/showthread.php/334967-Distribution-of-Star-Types-across-the-Galaxy

 

[_RUKUS_]

Going 3D sounds interesting, looking fwd to that. Here are a few 2D-views (incl. 26k ELW) from a while back: https://forums.frontier.co.uk/showthread.php/334967-Distribution-of-Star-Types-across-the-Galaxy

Amazing work Redfox! I am curious about the process you took to make these

 

[Redfox]

Amazing work Redfox! I am curious about the process you took to make these

Thanks! Only limited amounts of black magic were used: The EDDB-exports (bodies.jsonl, systems.csv) were loaded into a DB, followed by a query per star-/body-type (returning their coordinates), and finally that data was put into Excel scatter charts.

 

[_RUKUS_]

Going 3D sounds interesting, looking fwd to that. Here are a few 2D-views (incl. 26k ELW) from a while back: https://forums.frontier.co.uk/showthread.php/334967-Distribution-of-Star-Types-across-the-Galaxy

This is a crosspost, but I'm really not sure where to put it anymore, so apologies if you've already seen.

I finished up the 3D map using CMDR Biobob's amazing platform, and after having to rebuild my model from scratch I present to you: Earth-likes in 3D https://ed-elw3d.firebaseapp.com/
It now contains ~36,800 42,457 points (all systems containing ELWS, as of posting today). Info has also been added. To remove the background and display only clickable points, "Toggle Galaxy Map" under the cog in the upper right hand corner.

More features to come in the near future.

o7

 

[Ian Phillips]

Open sesame

 

[Brett C]

Bumping by request of op.

 

[marx]

Thanks! Didn't expect to get a Frontier stamp on the thread too.
So, for the update itself: as promised in the ringed thread, I've done a full update, so posting that here now. I've left the yesteryear version up as well, it might make for some interesting comparisons. For example, to see how things have changed in different sectors.
Also, this time around, ringed Earth-likes are specifically included as well, and new maps have been made with the Ringed / Earth-like ratio. I didn't make these for the top and bottom layer, because there are so few ringed Earth-likes there that it would have been pointless. For instance, in the bottom layer, there are only two.
I recommend making a copy for yourselves, and playing around with the cut-off values on the "Data" subsheet.

One thing to note: although I did still include the subsheets for summing up the Wregoe and Synuefe sectors, with the name-overridden mini-sectors as well, I made a mistake back then. I forgot that some of those overlap sector boundaries, and in such cases, everything would be counted twice, even if they are actually in another sector. While it would be possible to use the coordinates (where available!) to separate them into their actual sectors, that would be a lot of work, and I don't know if the sectors around there are different enough that it would be worth it. Plus we suspect that generation might differ not just in sectors, but down to boxels too.

Full planetary characteristics (like the second sheet in the OP) and coordinates will come later, I'll just have to see if I can massage them into Google Sheets in a similar way to what I posted here. If not, then I'll just share it in an offline document.

So, what changed? Besides the ELW sample tripling, and the total systems nearly doubling, that is. With more systems around both the bubble and Colonia, as well as along the Sol - Colonia and Colonia - Sag. A* lines, I'm more inclined now to say that the core might have better chances for Earth-likes. Still not quite sure about that, but it appears to have gone more that way since the last one. If this is true, then it makes the difference in ringed Earth-likes in- and outside the core even more interesting. Consider this: if(!) there are better chances for Earth-likes in the core, then how come there would be worse chances for ringed Earth-likes in there? Mind you, that's still a big if. At the risk of sounding like a broken record, more samples are needed.


Lastly, a fun bit I've noticed. There are almost four hundred sectors where the number of systems logged is below a hundred, yet 1-4 Earth-likes have been found. I looked up a few of them, and saw that most were first discovered by... well, I think you can guess (Or, even better, look it up yourself.) But if you don't want to,

Spoiler

a certain CMDR Allitnil.

 

[marx]

So, I ran things and checked over the results, and added the resulting list of Earth-likes to the OP. (Or here.) It includes coordinates for all of them, as well as the best I can do with Google Sheets. If somebody plotted those on an actual galaxy map, I'd be grateful. Comparing the "total" map with the "ringed" map, I think it illustrates quite well my earlier point about ringed Earth-likes, stellar density and the core. (See here.)

Note that I didn't include the planetary characteristics. (Surface temperature, gravity et cetera) I forgot that Orvidius' processed dump left some out that were in my earlier sheet, but looking at the ones that were available, the statistics stayed almost exactly the same since last year. I can still share them if somebody would like, but I don't really see much point in uploading it, nor in compiling the full data again. The conclusions there would be the exact same as last year.

Oh, one thing that was new: compared to the 20 mass code H Earth-likes that were all ELMs, now there are 58, of which two are not moons.

 

[Imperios]

Lastly, a fun bit I've noticed. There are almost four hundred sectors where the number of systems logged is below a hundred, yet 1-4 Earth-likes have been found.

I noticed that the same appears to be the case in the ringed ELWs sheet excel list that was posted here and wondered: Is there maybe a pattern behind the location of these sectors (=there are sectors with much higher chance of (ringed) ELWs) or has Allitnil just a golden nose finding them ?
At least it seems to me that there are far too much sectors with " low total systems discovered vs. (ringed) ELW count" to just be a lot of lucky cases.
Maybe it's all in sectors with very low star density and stellar forge generates a minimum amount of (ringed) ELWs per sector so chances in these sectors increases ?

 

[marx]

There is a pattern of sorts: they tend to be on the extreme edge of the galaxy, where Allitnil traveled a lot. IIRC there are mostly code D systems there, so going through those would explain that quite nicely.
As for the ringed stuff... Not sure. I'd be surprised if the Forge were set to generate a minimum amount of body types in sectors, but given how complex it might be, who knows? In places were there are very few systems, we can only guess.

 

[Imperios]

Just thought the answer to that might be way simpler: Allitnil did you maybe skip scanning most systems with no interesting bodies in the outer sectors ?

 

[marx]

Just thought the answer to that might be way simpler: Allitnil did you maybe skip scanning most systems with no interesting bodies in the outer sectors ?

Oh, I missed this line, and he didn't answer. He needn't have, though: systems are recorded even if you didn't scan anything in them. (Not surprisingly, the majority of systems have no scans.) The analysis here uses total visited systems throughout the sector counts.