Something I've pondered since I converted my XE-1 to infrared has been the possibility of capturing infrared pictures in the moonlight. Most pictures in infrared are taken during the daytime. I've only been able to find a few cases where infrared has been used at night-time and a few rare cases where it has been used with the moonlight (example here). The problem is most infrared shooters use an external infrared filter to capture infrared pictures. Doing this is a cheaper and easier option compared to buying another camera and spending even more money to convert it. The problem with this method is that it requires very long exposure times; usually somewhere between 5-30 seconds and this is in full daylight. At night, when the available light is decreased dramatically, using an infrared filter in front of the lens would require exposures far too long to be practical. Given that, I was still curious to see what infrared photographs would look like in the moonlight.
Before I continue I should dispel a common misconception that infrared cameras are somehow 'night vision' cameras. For the most part the same sources that create visible light also create infrared light and conversely a lack of visible light comes with a lack of infrared light. When people associate infrared with night vision, they are usually thinking of enhanced spectral range or active illumination. Enhanced spectral range just means that the exposure is capturing more than just the visible spectrum of light. This either includes more ultra violet or near infrared light which is useful for night vision because it allows more of the light spectrum to be amplified (more signal). Night vision googles, scopes, etc. like those used by the military use visible and near infrared light for this very reason. Active illumination involves shooting infrared with a source of infrared light to expose the image properly. This is just like using a torch or a camera flash to take pictures in the dark, but it would be a misnomer to call this night vision. The reason why infrared is associated with active illumination/night vision here is because humans and animals can't see infrared light. If you've seen a nature documentary you might have seen footage in black and white at night of animals going about there business. This would have been done by using an infrared camera with active illumination to expose the picture without disturbing the animals. The military also uses this when coupled with their night vision equipment for similar reasons, except this time not to alert people.
Moving back to the main topic, I was still curious how infrared light would behave at night without active illumination. At night the moon reflects light from the sun back to earth and this small amount of light can be used to take interesting pictures. Those familiar with the characteristics of infrared photography know that the reflecting properties of visible light often different dramatically from infrared light. Foliage for instance glows white with infrared white whilst appearing the normal green in visible light. The effect is due to the dramatic increase in the reflecting properties of leaves beyond 700nm although this can vary between species. The curiosity of mine centered around whether moonlight might differ in any way that would impact photography. If the moon didn't reflect enough infrared light for instance, it would make night photography difficult. So I decided to test this out for myself.
The first difficulty of photographing in the moonlight is you actually have to find a night when the moon is out. The moon moves in phases where it is more or less illuminated which in turn influences how much light is reflected back to earth. Ideally the time for moonlight photography would be at a full moon, when the moon is brightest. This narrows the available shooting days to around 5 out of the moon's 30 days cycle. Coupled with the chance of cloud cover it can be difficult to get good conditions to even start taking moonlight photographs. A month ago I managed to find a day when the full moon was out. I grabbed my equipment and headed up on of the local hills to take pictures of the mountains in the moonlight.
This first trip didn't turn out well. I had two troubles to do with camera focusing and a light leak. At night there is very little light and looking through the camera's viewer finder reveals a pitch black image. This isn't a problem for exposure because you can do a simple trial and error to narrow down the correct exposure. The big problem is that you can't see what you are taking a picture of and whether of not it is in focus. In this case I was using a manual focus Super Takumar, but if I were using a auto focus lens I would still have to have manual focused because of the lack of light. With visible light you can use the focus markings of the lens to find infinity focus, but because infrared light focuses closer than visible light (a topic I will definitely cover later), that was inaccurate. The Super Takumars are nice in that they have infrared focusing marks, but these were thrown off by my use of a focal reducer. Focusing then was hit and miss as I tried to slightly adjust focus between shots to find the correct position with little success. The focal reducer also created a light leak. I double adapted my m42 lens onto a EF to Fuji X mount Zhongyi focal reducer which works great for normal photography. Double adapting like this isn't ideal and this situation showed off why as the small gap between the m42 to EF adapter and the focal reducer was letting in a some light. This light was showing up in the pictures because the exposure time was so long (30s). Looking at the top left hand side of the picture shows off the light leak. The rest of the image is off focus too.
Despite this setback a month later when the moon returned I set out to try again. This time I traded in my Takumar with a focal reducer for a Pentacon 50 1.8 with a dumb adapter (no elements inside). Removing the focal reducer would remove the source of the light leak, but it would also slightly lower the light getting to the sensor (-~2/3 of a stop). This wasn't much of an issue because I could simply increase the ISO to compensate. The other important thing I did was to put some masking tape over the focusing ring and the rest of the lens when the lens was focused to infinity. I then cut the tape down the middle so the lens could be focused normally. When I needed to focus to infinity in the dark however, all that was required was to line up the two parts of the masking tape. (I switched to the Pentacon from the Takumar because the grip on the Takumar would have been in the way).
So I set out of a cold winter's evening to get some photos in infrared and here are the results.
Pretty interesting stuff. All these photos as you'll notice are black and white conversions. Doing false colour infrared conversions adds some interesting detail to the mix.
Also so very cool stuff. I should mention though that these pictures were taken not long after sunset. This meant that there was still sunlight coming from behind the mountains. This made colour separation as easy as the normal process. Taking pictures that didn't face the mountains found issues with colour separation. If you look at the following image that faced south rather than west there was some blue separation near the horizon, but further up the sky it started to turn red.
This phenomenon started to occur in the west too as the night went on. The last photo I took which faced west exhibited the same problem.
This shows that the colour separation ability of infrared was deteriorating. Although red and yellow are different colours, in the separation they are both part of the red channel (720nm really only allows for 2 colour channels, blue and red). What we are seeing is that the infrared light is becoming more uniform as the sun retreats and the moonlight takes over. It might suggest that there is a difference in the infrared light that the moon is sending out compared to the sunlight. This is curious and something I will look into further.
One other thing of note is that shooting wide open (in this case at f/1.8) looked quite nice even though sharpness fell. I think the soft dreamy quality of the focus suits the soft moonlight quite well. (Note: than the exposure has been changed in post so the ISO measure is approximate)
To conclude it is obvious that you can get interesting photos shooting infrared in the moonlight. In addition shooting wide open creates a nice dreamy effect in the moonlight and false colour infrared becomes difficult as remnants of the sunlight wane. What still isn't clear is why this second phenomenon is occurring. A possible answer is that the moon is reflecting light in a special way, but I can't be sure. I have not doubt in the future I will find myself on top of a hill again trying to answer that question. For now though, this will remain a mystery.