Hacker News Reader

Comment by haarts | Hacker News Reader

It's interesting to note that the highest ESI so far has been Kepler-62e with a score of 0.83 (https://en.wikipedia.org/wiki/Earth_Similarity_Index). So a score of 0.98 would seem significant.

View on HN

That Wikipedia page doesn't appear to be updated very often. Here (http://phl.upr.edu/projects/habitable-exoplanets-catalog/dat...) is a better list. KOI-4878.01 has been listed as a 0.98 since January 2015. The new may be that it is only now confirmed; but there are other confirmed KOIs with fairly close values of 0.93 and 0.91 already.

https://en.wikipedia.org/wiki/KOI-4878.01

Interesting comment by `admiralrewd` regarding KOI-4878.01 & general critique of ESIs on reddit from 6 months ago:

http://www.reddit.com/r/space/comments/2rbjo9/if_confirmed_k...

edit: the post links to a couple of papers making a few pretty reasonable points about the assumptions underlying the ESIs, and the errors involved in the calculation.

Question: why is 'escape velocity' part of the Earth Similarity Index when, presumably, it's just a function of radius and density (which are also in the index)? I feel like I'm missing something.

It's a reasonable measure of surface gravity, which, along with an idea about the distance from the star it's orbiting, gives us a good idea about what kind of atmosphere we can expect (i.e. the lighter and the hotter a gas is, the easier it can escape a planet with a particular escape velocity), at least a basic idea about what kind of surface features we might expect and so on.

It's also a parameter that is quick to compute starting from radius and mass, which are often the first things you know about an exoplanet due to the way they're found.

Right, but the escape velocity is ~ radius * sqrt(density) anyway, and if you look at the equation for the Earth Similarity Index[0], each term has an associated weighting. So why include escape velocity explicitly, rather than just changing the weighting of radius and density? The weightings seem to be quite specific numbers too, so I imagine there's a methodology behind it, I was just curious as to the logic. (Also, e.g., why is the temperature weight 5.58? Wikipedia says the weighting is to 'equalise their meanings', but it's not clear to me what that means.)

[0] https://en.wikipedia.org/wiki/Earth_Similarity_Index#Formula...

I think the formula that Wikipedia gives isn't used that often for most exoplanets. Oftentimes all you know about a planet is the mass, radius and some information about temperature or at least energy flux to its surface, so they use those at first.

I guess the parameters were chosen by some curve fitting procedure to make sure it gives 1.0 for Earth, 0.7-ish for Mars, 0.2 for Uranus and Neptune which are basically the most "not like the others" in our solar system and so on. It's not really a measure of habitability or anything like that. "Equalizing the parameters' meanings" presumably has to do with mapping their relative variation (and the effects of that variation over various observable parameters) over a smaller range.

I'd take it as a proxy for surface gravity. They can't exactly say what the mean surface gravity is, because they don't know how much variation there is in altitude, and whether the exoplanet has a sea level.

Even so, giving a rough estimate of mean surface gravity would make the tables more accessible to non-experts.

Reply on news.ycombinator.com