Exposure for capturing moon and landscape

[walang_pangalan]

As noted above, there are relatively rare situations where one exposure is enough. But in most cases, multiple exposures are the only rational approach. Evangelical kooks who preach that "compositing is a satanic manipulation for which you will BURN IN HELL FOREVER" are free to proselytize, I suppose, but the reality is that a few frames over a few seconds -- exposures tailored accordingly -- isn't much different from a single exposure that lasts a few seconds. Photoshop even has an HDR tool that is specifically designed to solve the problem in a (hopefully) principled way.<div></div>

[jeff_conrad]

I don't think anyone here was suggesting eternal damnation for any kind of

compositing, even if the resulting image is complete fantasy. Exposures

taken a few seconds apart to handle the contrast barely even qualifies as

compositing as far as I am concerned, and, in essence, it differs little

from what was done in the darkroom for years. Even a bit of cleanup has

been standard practice, though it sometimes has been criticized. Adams

originally got a lot of flak for removing the "LP" from one of the

foreground hills in his <cite>Winter sunrise, Sierra Nevada</cite>; this

strikes me as a bit extreme. I suppose a purist would have rounded up the

Lone Pine High School students responsible for the lettering, marched them

up the hill at gunpoint, and made them clean off their handywork, but I

think Adams made a reasonable decision to do the cleanup in the darkroom

<p>

There still is good reason to get a Moon shot with a single exposure,

however--it's often a lot easier. The choice obviously depends on shooting

conditions; with a clear sky and a sharply defined skyline (e.g., a moonset

over Mt. Whitney), compositing often <em>is</em> a ten-minute operation.

However, if the sky contains clouds, or the Moon is juxtaposed with a

landmark, as in Walang's image, the Moon's motion can make registering the

two images (or even portions thereof) a bit dicey. If it's truly a

nighttime shot (i.e., more than about 30 minutes after sunset, compositing

is the only choice, and if the sky is clear and the Moon is above anything

on the skyline, compositing again is an easy operation. I don't take many

photographs of this type, but when I do, I try to get one exposure before

the Moon appears and another (or several others) when the Moon is in the

desired position.

<p>

If the first exposure is made after dusk, the result corresponds to an

event that was actually observable. It's a bit fuzzier when the first

exposure is made during twilight: the combination of Moon and sky condition

didn't actually happen. However, even this isn't much different from what

long was standard practice for evening architectural photography: an

in-camera double with the first exposure made 5-10 minutes after sunset to

get the sky, and the second made after the lights come on.

<p>

I probably get at least 90% of my shots with a single exposure. I cannot

recall a single image that I missed where the cause was other than weather

(or perhaps an occasional bad aesthetic decision). Undoubtedly, there are

opportunities that I miss (or don't even attempt), but since I usually try

to get the Moon between 5 and 20 minutes after sunset (or before sunrise)

for the sky color as well as exposure balance, I'd miss many of those

opportunities anyway. I'm often out with others (some shooting film, some

shooting digital) who take the same approach. We're not all fanatics ...

<p>

I often continue shooting past the prime of the event, and at that point

always make at least two exposures for every composition. In the vast

majority of cases, my selects have been images made with single exposures.

<p>

I've yet to try HDR for a Moon shot; I'm not even sure it was intended for

this type of image, though I'll probably give it a try. I anticipate one

difficulty: when blending two complete images, the Moon's motion still may

be an issue. I always lock the mirror when using a long lens, and given

that, I'm doing well if I can get two exposures within 6-8 seconds of each

other. Maybe I just need to practice ... If the motion issue can be

managed, HDR might provide a means of handling the contrast with a few

other bright elements (e.g., clouds or even street lights) in a scene.

<p>

The original poster asked if there was a way to manage the contrast, and I

simply tried to show that more often than not, this can be done. There

certainly are other ways to get a landscape with a Moon, and they aren't

necessarily less valid than my approach. On the other hand, I see nothing

wrong with using a bit of planning to get the shot without resorting to any

significant manipulation.

<p>

Incidentally, I'll second Walang's comment about the infrequency of

repetition; the frequency obviously depends on how strictly one defines

"repeat," but in most cases, a combination of Moon position, phase, size,

and time never repeats. The weather is yet another issue ... For example,

the following alignment won't happen again in the physical world until 5275

(and even that won't be exact). This was a single exposure, but it took

some work to preserve the texture in the Moon. I put the sky on Zone IV

and misunderestimated the Moon-sky contrast. I didn't make a second

exposure because I didn't think one was needed (okay, I don't always get

things to perfectly balance ...).

<p>

The recurrence in 5275 assumes, of course, that Jim doesn't take the

building out in the interim with a botched fill flash.

<div></div>

[greglyon]

Here's one I did last week. 2 exposures. Photoshop. All i did was refocus (quickly!) between the two exposures. No 500mm really has this kind of DOF! <a href="http://www.photo.net/photo/3481197&size=lg">Link to larger size image.</a><div></div>

[greglyon]

Oops, forgot the point: This was an event that I actually witnessed that my camera/lens could not capture with a single exposure. And to allude to another thread, no, I didn't check the 'unmanipulated' box when I posted it!

[kelly_flanigan1]

Both the Earth and Moon are illuminated by the same almost point light source; the sun; the same distance away. Go grab a Volleyball; and a Golf ball; and stand under a street light; to "discover your answer". This is a lighting ratio issue; and exists even beyond still photography; in dance; theater; the MOVIES, the kings of lighting experience. Just grab the two balls; study the lighting of them; as the golf ball rises above the volleyball; with a distant light source. Add some smoke is you want some atmosphere. In ancient film school the thought was to studdy lighting's effects; and not get bogged down in details that didnt matter. Shooting real movie film is expensive; studying lighting is where it is at.

[frank uhlig]

G,

 

your quest can be solved by spotmetering (A) the rising moon that needs exposure by the f/11 rule, maybe more in light clouds, and (B) the landscape on earth.

 

If these light values differ by no more than 5 light values or whole stops, you have a slight overexposure/underexposure on your film that you might be able to save when printing. If the light value difference is 10, as it occurs quite naturally, you have no prayer with one shot, no PS.

 

So, spotmeter the earth landscape since you know the moons' light value by heart and see and wait for good days! Of ourse it helps to have an illuminated earthscape such as a city skyline rather than a nuclear bomb to light up your back ...

[jeff_conrad]

Measuring the luminance difference with a spotmeter certainly is the

desirable way to see if you have a contrast problem; however, it's not

always easy to meter the Moon. The Moon's angular diameter is 1/2 degree,

so a 1-degree meter won't do the job; however, with a long enough lens, an

in-camera spotmeter can make the measurements if the Moon is completely

visible. If part of the Moon is occluded, the spot metering area may be

larger than the visible part of the Moon, and the measurement isn't

reliable.

<p>

In theory, it's possible to estimate the Moon's luminance from measurements

with a 1-degree spotmeter. The Moon occupies roughly 1/4 of the metered

area, so the actual luminance should 2 steps greater than indicated. I

have known some people who say they have had good luck with this approach,

but it hasn't worked for me with either a Pentax V or modified Pentax

digital. If the sky surrounding the Moon isn't completely dark, the

estimated metering approach breaks down completely.

<p>

I've never understood the f/11 ("Loony 11") rule, because the Moon's

luminance changes with altitude as its light passes through different

amounts of Earth's atmosphere. The difference is substantial: as much as

10 exposure steps between the horizon and an altitude of 10 degrees; at

which altitude does the f/11 rule apply?

<p>

If I could independently control it, the "proper" exposure for the Moon

would depend on the desired rendering as well as the Moon's luminance. The

Moon is a fairly dark object, about 10-12% reflectance near the time of

full Moon, but subjectively, it usually is brighter. Even the perceived

brightness varies with altitude and sky condition. Near the horizon, a

Zone V (or perhaps slightly lower placement) might match my subjective

impression; higher in the sky, a Zone VI rendering probably would be a

better match. If the Moon goes much above Zone VII with transparency film,

most of the detail is lost; you might hold detail in Zone VIII with

negative film. Additionally, no two moonrises are quite the same; the

"best" rendering, especially near the horizon, is a matter of personal

preference that doesn't necessarily follow any rule.

<p>

With a single exposure, of course, the Moon's exposure isn't independent of

the exposure for the foreground. Most of the seemingly endless discussions

about the exposure for "the Moon" seem to forget that, absent a big,

honkin' lens, the Moon occupies a very small portion of a landscape

photograph, so to have any picture at all, the exposure for the landscape

needs to be right. This isn't necessarily the meter indication; I

typically meter the sky away from the Moon, and place the sky on Zone V

just after sunset, and gradually reduce the placement to Zone IV about 20

minutes after sunset. This is my approach and YMMV; my placements

typically are 1/2 to 1 step darker than the actual subjective impressions

of a scene.

<p>

If you get the picture on the day of or before the full Moon (the time

difference between moonrise and sunset is the real criterion), it's often

possible to get an acceptable exposure for the Moon when exposing the

foreground as I described. When the luminance difference exceeds the

maximum (with slide film, about 3 steps), a separate exposure for the Moon

is needed if you want to hold detail.

[kelly_flanigan1]

When the moon is high in the sky; and full; it is flatly lite by the sun. Then the exposure for the moon is the same as the F16 rule; ie 1/asa @ F16. With my F8 reflector; I expose about 1/1000 at F8 with tri-x; or 1/250 @F8 with Plus-x for the full moon. Both the moon and earth are illuminated by the same light source; the sun; 93 million miles away.

[jeff_conrad]

The Moon indeed is illuminated by the same light source as the Earth, and

were it not for Earth's atmosphere, the f/16 rule would apply. Of course,

were it not for Earth's atmosphere, we wouldn't be here to have this

discussion.

<p>

Years ago, the Nikon School of Photography made the same statement, and I

believed them until I actually tried it some years later. With the Moon

near the horizon, it simply doesn't work. At the horizon, the Moon's light

passes through approximately 38 times as much atmosphere as it does at

zenith, and the attenuation of that light is substantial. Demonstration of

this effect is simple: pick a night when the Moon rises 45-60 minutes after

sunset (often, the day after a full Moon), use a moderate telephoto lens

(85-135 mm on a full-frame 35 mm format), and make exposures of the Moon

every 5 minutes for an hour or so, using the same settings for each

exposure (I'd suggest 1-2 steps greater exposure than the f/16 rule). It

will be quite apparent that the Moon is redder, darker, and more distorted

at the horizon. It also will be apparent that the Moon's size at the

horizon is the same as when it is high in the sky (actually, it gets

slightly larger as it gets higher).

<p>

You won't see the effect of atmospheric attenuation as easily if the sky

isn't dark; the light from the Moon and the sky combine. When the Moon is

much darker than the sky, it simply gets faint, although it often can be

observed because of the color difference. Greg's photo illustrates this

quite well.

<p>

When the Moon is high is the sky (say, 40 degrees or greater), the f/16

rule works much better, although there still is some atmospheric

attenuation. The result usually is a low Zone IV placement; for a picture

with a long lens or through a telescope, in which the Moon is the only

element, this well may be what is wanted. If part of a landscape, however,

this rendering may not match the subjective impression of the Moon's

brightness. For a more extensive discussion of this, see Ansel Adams's

<cite>Natural Light Photography</cite> (Basic Photo Four. Boston: New York

Graphic Society, 1952), "The Moon and Moonlight." One caveat: I'd ignore

the claim that the Moon's luminance is 250 candles/square foot; this may be

true when the Moon's altitude is 40 degrees or above, but it's not true at

the horizon. Aside from this, Adams's gives what's arguably still the best

treatment of the subject.

[kelly_flanigan1]

Jeff ; my statement was <i>"When the moon is <b>high</b> in the sky"</i>. <BR><BR>In astronomy this means NOT at the horizon. <BR><BR>Here I have used the F16 rule with my f8 telescope; before the Nikon School existed; with an Exakta VX. With my scope; if I add one more stop; the moon will be correctly exposed; <i>"When the moon is <b>high</b> in the sky"</i>.<BR><BR>Here I have shot many many thousands of images of the moon; with many different telescopes; afocal rigs; with binos. I am well aware that the sunrise and sunset passes thru a longer optical path. In one total lunar eclipse in the 1970's; I shot about 5 dozen rolls of kodachrome; over the many hour affair; using a motor guided telescope. I am not sure why old Ansel is brought up; I have probably shot 1000 times the number of moon images. I sepent a few summers with an observatory; and shot film each night; and processed it each morning. When one bulk loads microfilm in 35mm; and shoots 1000's of images; the exposure tends to get easier to figure out.

[bobatkins]

<em>When the moon is high in the sky; and full; it is flatly lite by the sun. Then the exposure for the moon is the same as the F16 rule; ie 1/asa @ F16.</em>

<p>

True if you want the moon its true color, which is dark grey. Most people want the moon to look bright, and so the usual rule of thumb is to modify the f16 rule to f11, i.e. 1/ISO at f11. <p>

The following image was shot at 1/60 @ f10 using ISO 100. That works out to the same as 1/50 @ f11 or 1/ISO @f8. It's not <em>quite</em> full moon though. No level or curve adjustments have been made and the image was converted from RAW with no exposure compensation.<div></div>

[jeff_conrad]

Kelly, if you read my statement, I don't think I really disagreed with

yours. However, the original question was about photographing the Moon

rising above the horizon, and I wasn't sure why you suddenly switched to

astrophotography. It's even less clear how a total eclipse fits into this

discussion. I meant to say that it's a mistake to apply

astrophotographic exposure rules to landscape photography, as the Nikon

School (at least the one that I attended in the late 1970s) apparently did.

I made the distinction between the two types of photography quite clear in

my article that I referenced. I probably could have stated it more clearly

in this thread.

<p>

Why the mention of Ansel? As I said, aside from ignoring the effect of

atmospheric extinction, the work I cited remains one of the best treatments

of photographing a <em>landscape</em> that includes a Moon, including

rendering of the Moon as it usually is perceived in that context.

<p>

The best reference on exposure that I've seen probably is Bryan Peterson's

<cite>Understanding Exposure</cite> (New York: Amphoto, 1990):

<blockquote>

After doing extensive research on lunar-landscape exposures, I learned that

they should be treated no differently from a frontlit landscape without a

moon. ... during the first fifteen minutes immediately following the

sunset, the light in the eastern sky has the same value as the landscape

below. As a result, lunar landscapes can be metered in the same way

frontlit landscapes are.

</blockquote>

<p>

Kelly, Bob, and I really are saying the same thing, and I assume that we

all understand the difference between astrophotography and landscape

photography. The distinction may not be obvious to everyone, however.

Photographers often make one (or more) of the following mistakes when

photographing a landscape with a Moon:

<ol>

<li>

They make long exposures that tend to burn out everything; or,</li>

<li>

They use the f/16 (or the f/11 or whatever) rule when the Moon is near the

horizon, and get severe underexposure; or,</li>

<li>

They wait until far too long after sunset, when the Moon-foreground

contrast is unmanageable. The higher the Moon is in the sky, the worse the

problem becomes.</li>

</ol>

Sometimes, the f/11 rule will appear to work near the horizon, but it's

more usually because that coincidentally happens to be the appropriate

exposure for the landscape.

<p>

Because this thread has been about landscape photography, I wanted to

stress that astrophotography rules don't apply. Again, I probably could

have made this intent more clear.

<p>

Perhaps we repeatedly invite this confusion by asking, "What's the correct

exposure for the Moon?", when we really should be asking, "What's the

correct exposure for a landscape that includes a rising full Moon?" As

Peterson probably would have answered, "Pick a time shortly after sunset,

and expose for the landscape."

[g1]

My response to Kelly and Bob's more recent input has been addressed and expressed very well by Jeff. Indeed, the question posted was not intended to cover astrophotography [which is a very specialised field and not where I was intending to venture]. The first sentence of my question was; <b>I would like to capture a picture of the moon rising from the horizon and set above a foreground landscape.</b><p>

 

Walangs comment; <b>Evangelical kooks who preach that "compositing is a satanic manipulation for which you will BURN IN HELL FOREVER" are free to proselytize</b> is imo way off topic and completely irrelevant, and I would like to clarify that nobody has preached anything! All posters in this thread who have attempted results in-cam have supported PS methods as an option, yet <i>prefer</i> to take the in-cam approach [multiple exposure or one-shot].<p>

Thanks to all posters for their interesting and informative contributions. It's been an education!

[bobatkins]

I think the difference between a PS composite and a dual in-camera expsoure is very small from an "ethical" viewpoint. In both cases, the image is not a "true" representation of "reality".

<p>

I can see wanting to do it in a single shot, simply because it's more difficult and more "pure" and so more of a challenge. It's like someone wanting to walk from New York to San Francisco rather than drive or fly. That's they way they want to do it and it makes them happy. From the point of view of the outcome (getting from A to B) it makes no sense, but maybe they are more interested in the process than the outcome, even if nobody else cares.

<p>

The original question said: "<em>...then decided a double exposure would be the best option, but I am using digital...</em>, to which I'd reply "so what". There's zero difference (technically or "ethically") between a double exposure on the same frame of film and combining those same two exposures digitally

[jeff_conrad]

When doing astrophotography, one <em>should</em> understand appropriate

astrophotography exposure rules. Someone wrote:

<blockquote>

When the Moon is high is the sky ... the f/16 rule works much better ...

The result usually is a low Zone IV placement

</blockquote>

Uh, wouldja believe "a low Zone V placement"? The nonsense that one

sometimes encounters on Internet fora ...

[jeff_conrad]

<cite>It's like someone wanting to walk from New York to San Francisco rather than drive or fly.</cite>

<p>

C'mon, Bob, a true pilgrim would make the journey on his hands and knees ...

When he arrived, he would be astounded to find other people who had made

and continue to make the same journey, such as

<a href=http://www.moonchasers.com> James Rigler</a> and

the SF Chronicle's <a

href=http://www.sfgate.com/gallery/buyphotos/larsonsmoons/> Fred Larson</a>

(he even gets a few eclipses).

<p>

As I've said, I'm not convinced that it's that difficult to get the image

in one shot, though you need some idea of what you are doing, and you

definitely have fewer opportunities. I certainly agree that there is no

ethical difference between an in-camera double and a similar composite use

PS. I've even suggested that there is little, if any, ethical difference

between a single shot and a composite using two images taken within a few

seconds of each other, whether or not it qualifies as unmanipulated.

<p>

There still are quite a few people who prefer to depict actual events,

whether or not it's the most expeditious means of getting from A to B

(though sometimes the path is constrained by one's definition of B). I

don't suggest that they are aesthetically or ethically superior, but simply

that they don't necessarily merit consignment to the Loony bin.

[g1]

Bob, the methods of double exposure on film, and compositing two digital shots are not an ethical question for me [i use PS for many illustrative projects]. My reason for not wishing to use PS for this moon shot are a) because I already know how to achieve results with PS. I don't on the other hand, have much experience with double exposures on film [re the moon] or capturing the moon in one shot. The excercise then, is to push myself and to learn. And b) given the choice [i am doing it for myself], I would rather spend more time and effort with my cam in the field than on my PC at home. <p>

I would feel more proud capturing in-cam rather than with PS, because in PS I know I would be re-positioning the 'best' of the two moons, lifting shadows, and generally doing it the 'easy' way. If I were using a double exposure in-cam I would have to learn about capturing the shots so close together so as not to have two overlapping moons. I am not sure how the exposures would be managed in order to avoid that. But sadly, this isn't an option as I sold my two film cams. So, one-shot is what I would like to capture and I think advice about the moonrise/sunset times are critical, and responses here have been very helpful on that score. <p>

I agree about a wider aperture when the moon is on horizon rather than higher in the sky. When I tried my first horizon/landscape shot the moon was very orange and contrast was much lower compared to the brighter clearer moon when it had risen only half an hour later.

[jeff_conrad]

I don't see how you could handle this with an in-camera double using

exposures made in rapid succession. With in-camera doubles, exposures

always are additive: if the contrast was too much for a single exposure,

the foreground exposure would burn out the Moon, and any additional

exposure simply would make it worse. It would be much easier to pick a

night when the Moon rose half an hour or so after sunset, make the

foreground exposure during early twilight, and wait for the Moon to move

into view before making its exposure. By then, the sky would be dark, and

the only part of the second exposure that would contribute to the final

result would be the Moon itself--nature's best approximation of a blue

screen. You'd obviously need to have a good estimate of the Moon's path to

get a good composition. To me, this would seem far more difficult than

getting it in a single exposure. A digital composite would seem much

easier while every bit as "ethical."

<p>

With a digital camera (or scanned film), nothing says that you can't try to

get the shot with a single exposure, yet make a second exposure for the

Moon in case the contrast is too great in the first exposure. One thing to

keep in mind with a digital composite: unless the Moon is captured at

approximately the same altitude as where you put it in the composite, it's

not going to look quite right. This constraint, of course, is in addition

to having a believable lighting angle on the foreground. Making the two

exposures in rapid succession avoids both of these problems, whether the

timing is close enough that you simply can merge the two exposures, or

whether you cut the Moon out of the second exposure and paste it into the

first.

<p>

Of course, many people probably wouldn't know the difference between the

Moon's appearance at 3 degrees vs. that at 10 degrees, and probably even

fewer would care.

[ellie designs]

I'm probably repeating what has been said but will add my 2 cents... there is a window of time in which you can expose both properly without double exposure, before it gets too dark, but with a fairly fast shutter speed in my experience, and as you said the moon rises fairly quickly. I've done 3-4 moonrises in my lighthouse series and have experimented a bit with it. I'm assuming you've done this already, just look up the time for the moonrise wherever you plan to shoot, and know the location and appearance of the moon relative to what you want to include, its position will change at different times of year. Double exposure is fun too.