What are the odds of a satellite re-entering over land?


I love this one! I tell people the Earth is about 75% water on its surface, so the odds favor a re-entering satellite burning up over an ocean somewhere (the point being the odds of someone getting hit are that much lower). But is that really true? Not all satellites come down along the same path, and that changes the situation. An orbital debris specialist does the math.. and it turns out the answer really is pretty close!
Kevin Renfrow's profile photoWinchell Chung's profile photoChristine Forber's profile photoAmrita Mohan's profile photo
Chart is skewed because there's less area at extreme latitudes (90/-90). At the poles you either have 0% land or 100% land.

All things being equal, you're more likely to impact at equatorial latitudes because the circumference is greater.
+Mark Noble True, but you can account for that doing the math of orbital inclinations, which is how this particular chart was used. In the post I have the actual percentages plot too.
+Philip Plait I wonder how this plot would compare to a graph of known impact craters by latitude.
I not only knew that newsletter existed.. I've been reading and downloading for years.. except for the last several! Great source of information. Thanks for the reminder...
Interesting for quite a few reasons - just like the good old mercator projection, this graph vastly over-represents the size of antarctica and the north, and under-represents the size of the equatorial areas. Technically, as a strict interpretation of percentage, it may not, but the visual and relative effect is definitely mercator-esque
The amount of overrepresentation of high latitudes would be like a plate carrée or simple cylindrical projection, in which latitude-longitude coordinates just transform to a square grid. The factor in the Mercator would be the square of that, I think.
Is the latitude even a proper variable for this? I suppose that the orientation of the orbit (generally not parallel to the equator) would be of interest here. Two variables are needed for defining the orbit, even if the position on the orbit is considered to be stochastic and the height of the original orbit is not taken into account.

Most likely the orbits are not at all evenly distributed due to some being more practical for various purposes than others, the positions of launch sites dictating which are most economic and also for avoiding collisions by using a few standardized orbits. Someone with more interest on this should probably check with +NASA because I am simply guessing here.
That's a neat graph but how many satellite orbits parallel the equator? They like to show satellites pulling north of the Bering Sea, long course over the main mass of North America, pass quickly over the Atlantic, then presumably the long orbit above Africa and Asia. Something with proportions of landmass under the various standard orbits (inclinations?) might be more predictive.
Click through to the full article, +Lasse Kärkkäinen +Michael Rainey etc. There you'll see the percent landmass by inclination, I think presuming a random phase (longitude). The chart here is just part of the process.
Yeah, I didn't click through before but now I see it. That's really cool. Shows you'd have to try just to get one to land on your continent.
i want to see this with my own eyes "Orbital Debris Quarterly Newsletter" :)
for extreme eccentricity (Molniya) orbits, re-entry will be near perigee's latitude only, right ?
Add a comment...