How do they do that??

January 16, 2008 10:21 PM

How do you judge large horizontal or vertical distances visually, with no (or few) other objects around to compare as a reference?

The story about the UFO sighting in Texas made me wonder something, or rather, reminded me of something I've been curious about for a while now.

One of the men who claimed to see this craft in the air judged it to be 3000 feet or so above the ground. Others judged the dimensions of the craft in terms of football fields, or miles (yes I know, that's a big difference) in length.

My question is this: How is it possible to estimate, from ground level, the height at which an object is in the sky? (Never mind the further complicating matters of not knowing what the specific object is that triggered this question, nor its mystery dimensions -- I don't necessarily care to dissect this particular incident, it just got me thinking is all). Is there some trick to being able to figure it out, without any special instruments -- just your eyes? Even if it was something of known origin and size, such as an airplane or a bird. Perhaps not so much a cloud because they come in all sizes.

I am also curious how one would go about estimating the same but at ground level, e.g. a horizontal distance. I am horrible at this and wonder how it is so easy for others to at least throw a good guess out there. For example, let's say you're in a relatively open outdoor area, and you see a deer running across your field of view from left to right. There are some people who would be able to quickly ascertain that the deer was 400 or 500 feet away. How??!!

Is there skill involved? Really good intuition? Secret black project military training?

posted by brain cloud to grab bag (14 comments total)

It isn't possible.

But some people may think they can do it, and will claim they have.
posted by Steven C. Den Beste at 10:28 PM on January 16, 2008

There are some people who would be able to quickly ascertain that the deer was 400 or 500 feet away. How??!!

According to a guy I know, if they play golf regularly, they'll have more experience at accurately judging distances. Serious hunters may also have some skills.

I don't know how true that is - I've never played golf, or hunted. But both he and I volunteer as hawkwatchers, counting migrating hawks as they fly by, and I can say with some authority that it's really hard. Distance and light conditions can wreak havoc on one's ability to judge the size and color of something moving in the air. It can play games with your sense of how fast something is moving. Experienced hawkwatchers all have stories about the time they saw a large bird far away that turned out to be a smaller bird close in, for instance.

Because I hawkwatch in a place with known-distance landmarks, I have a kind-of sense of how far half a mile is, how high 300 feet is. But it's still very, very difficult. Add to that the fact that people will see what they want to see, and you should take eyewitness accounts like this with a huge grain of salt.
posted by rtha at 10:36 PM on January 16, 2008

How do you judge large horizontal or vertical distances visually, with no (or few) other objects around to compare as a reference?

1. Parallax. This is how your eyes do it although it doesnt work as well for really large distances.

2. Visibility. A closer object will be "clearer" while a distant one is more hazy.

The deer is easier since you have a conception of how fast a deer can run and can then estimate the distance by how fast it moves across your field of vision. This is more intuition than math.

It isn't possible.

What isn't possible? An outfielder can certainly catch a ball coming toward him from out of the sky with no field of reference behind it.
posted by vacapinta at 10:45 PM on January 16, 2008

I don't have any advice other than to go backpacking for a while. I seem to be better at estimating how far off a mountain is after spending a couple of months hiking. Also seem to be better at estimating the elevation of nearby hills. I guess after a while you just get good at guessing how much pain you are about to endure :)
posted by meta87 at 10:46 PM on January 16, 2008

One trick I learned as a kid used the Theory of Similar Triangles. My thumb is about 1" wide, and if I hold my arm out and stick my thumb up, it is about 20" from my eye. The ratio is then 20:1. One story of a building is about 10' high. For a one story house this would be the distance from the ground to the base of the roof, or for a multi-level building from window to window. If I hold my thumb out and it exactly covers the distance of one story (10') then I am about 200' away from the building. If it covers 2 stories, or one story is covered by only half my thumb, I am 400' away, etc.

This is good for something you have a size reference for at close distances. For longer distances, you need experience. Find a reference, like a water tower or airport tower and drive a mile away from it. How clearly can you see it. Drive another mile and compare.
posted by Yorrick at 10:47 PM on January 16, 2008

There is are a number of different visual cues for assessing distance. A good list can be found here. The most relevant in this case would seem to be:

Retinal Image Size - You may know approximately what size other things in the sky are and compare (e.g. a bird, clouds)

Texture Gradient - The farther away an object is, the less definition the texture of the surface has. Something that looks smooth is typically farther away.

Aerial Perspective - The farther away something is, the blurrier it will be because of particles in the way.


Skill could be involved in that familiarity with actual sizes of objects you can compare it to help. People also have different vision capabilities, etc. However, for an object in a clear blue sky that you are unfamiliar with, it is very, very difficult to accurately assess it's size and distance.
posted by christonabike at 11:01 PM on January 16, 2008

... not knowing what the specific object is

I think that's at the heart of your question. You have a reasonably good idea of how big a deer is, so its size relative to your field of vision can give an indication of how far away it is. For something like a UFO, without any reference points, it's impossible to say how big the object is - it could be huge and far away or small and close.
posted by sanko at 11:03 PM on January 16, 2008 [1 favorite]

With the deer, you can tell their approximate age and gender, and their size varies within certain limits. So if you know a fully-grown male deer is such-and-such high, you can judge from how small it is, roughly how far away it is. This is especially the case if you've used some definite reference in the past to familiarize yourself with distance from deer; a good example would be setting up your hunting blind a known 300 yards away from a tree stump, then watch to see deer come near it, and now you know how big a deer 300 yards away looks. It's the mathematical reverse of a depth of field algorithm.

Same applies to size/distance of people; ask an experienced football player how far away a person is standing, if it's less than 100 yards or so, and he'll probably be able to tell you within a yard or two, as he's spent so much time on a grid where the distance is marked out and matters. Professional surveyors, even more so. It's a heuristic ability; work it out with instruments a few hundred times, and you start being able to guess it by eye.

However with the UFO, even if you've no idea how big it is, you have some other cues - did it pass behind or in front of clouds? How "blurred" is it? (Distance can be guessed from air-blurring, to some extent.) You can safely guess its size within an order of magnitude - given that it is a dark object visible in the blue sky, it can't possibly be larger than ... some amount. (Ask an optical physicist.)
posted by aeschenkarnos at 11:05 PM on January 16, 2008

With unkown objects at a distance, you just can't get an accurate grasp on them. You see a cow standing on a featureless plain, you can make a good assumption based on how big you know a cow to be.

The best way to get better at guessing distances, as in your deer example is to practice, as it is a skill. I am pretty good with guessing distances up to a kilometer or so, because I walk with a GPS on my wrist and have spent time estimating distances to object as I walk towards them.

You see some lights in the sky, and you might think they're points of a massive object, far away, moving at trememdous speed when in fact they are closer and slower.

I read a comment, it may have been on MeFi or somewhere else in relation to this `sighting'. The poster said that one night he had seen three objects travelling across the clear sky, and he assumed they were something like satellites until they made an abrupt turn and he realised that they were moving at colossal speed and making impossible turns.

As he returned home, he saw them again, but as the objects passed into a more lit area, he realised it was an owl, just at streetlight level! He was seeing the bright patches on it reflecting the streetlights.
posted by tomble at 11:11 PM on January 16, 2008

Tangentially, I couldn't help noticing that two different witnesses interviewed on NPR yesterday morning both said the UFO was about 3000 feet up. Considering the difficulty of making such judgments about an unknown object, I got the feeling that someone had tossed out a number that got repeated by people who really had no clue.
posted by jon1270 at 4:10 AM on January 17, 2008

It isn't possible.

Of course it's possible. It's just the accuracy that varies with experience, knowledge and personal judgement.

I can watch racing cars being driven at high speed (100 - 130 mph) from quite a distance away (maybe up to 5-800m away), and tell you how the car is handling, and (often, but not always) if it is driver induced or not. The suspension movement of the car is in the realm of 10-20mm, so perceiving that sort of detail at that range seems (and is) impossible to someone that hasn't been doing this sort of thing for 20 years. Yet I can do it relatively accurately - proven by checking my conclusions by looking through the telemetry and getting the driver's feedback after the car has run.

Pilots are pretty good at judging heights and distances of airborne objects particularly. As is someone with reasonable knowledge of cloud and weather conditions. You can pretty easily work out how high clouds are once you know formations, and that gives you two relative points to work with. They aren't, after all, saying "exactly 3000 feet" but you could add +/-500 feet to that and you still have a rough idea of how high it was, even with that much inaccuracy.

You use perception of detail to discern how far away it is, and it's movement relative to objects in fixed positions. Of course, if the object really is unknown to you, then as you have no idea of the expected detail, inaccuracies (like in the hawk example) can be easily introduced by assuming one thing based on experience, when the object doesn't correspond to those rules. If, for an aviation example, you assume (even subconsciously) that a vertical tail fin on a plane is around 6 foot high, it will colour your decision as to how far away it is. Maybe you've just never seen a 20 foot high one. Or vice versa.

This is why realistic model aeroplanes (ie flying scale models) can fool people into thinking they are real and further away until the perform in a manner that jars with your perception of how that plane can turn/climb, etc. Then you get confused, and then twig it's a model.
posted by Brockles at 6:39 AM on January 17, 2008 [1 favorite]

It's certainly possible. I do visual estimation of distances as part of my job. It's farily easy to train someone to be 25% accurate with a few days experience. The best way I've found is to guess a distance then use something like a laser range finder to check. With a bit of practice, you get much better at it.
posted by bonehead at 7:43 AM on January 17, 2008

i heard a trick once ... cut the distance in half and see if you can estimate that. if not, repeat until you can estimate.
posted by maulik at 7:45 AM on January 17, 2008

It is definitely much harder to do with unfamiliar objects. I live in DC, and have had several conversations with folks about how surprised they are about the distances between the Capitol and the Washington Monument, or the Monument and the Lincoln Memorial. Their untrained eye usually under-estimates the size of these objects, so people think they're much closer together than they actually are--it's about 2 miles between the Lincoln Memorial and the Capitol.
posted by MrMoonPie at 9:37 AM on January 17, 2008