Cinematography Mailing List - CML

 

Search

 

Filming The Aurora Borealis

I'm shooting in Churchill in Canada in November and hope to capture the aurora borealis on film. I'm taking my Aaton and an intervalometer.

Anyone out there with helpful suggestions from their own experience?

Tony Coldwell



>I'm shooting in Churchill in Canada in November and hope to capture >the aurora borealis on film.

I'm sure you've thought of doing some research on the net for this information. One of the best sites on photography of Aurora’s is:

http://climate.gi.alaska.edu/Curtis/curtis.html

One of the more interesting sets of links found at this site are the one's that take you to aurora forecasting sites.

Cliff Hancuff
www.ClearDaySoftware.com
Publishers of SunWhere(tm) and MoonWhere(tm)
Washington, DC



Tony :

Try to check out a movie called Pictures of Light (if you can find it - it was in festivals around '95). Great doc about a film crew going up to Churchill to shoot the aurora borealis. Not only is the movie's goal to put that amazing phenomenon onto film, but it also ends up being a document of the crew's process of shooting such a difficult and elusive subject and the hardship they endured to do it.

Based on the film, I would say bring some real warm clothes, a good intervalometer, and a very long book.

Have fun,

Rob Barocci
DP - New York City



>Pictures of Light (if you can find it…shoot the aurora borealis

Wow! If anyone CAN find this film, please let us know where. I spent several weeks in Iceland in 1969 (a lifetime ago!) trying the same thing.

Got some good stills at least, but there just wasn't the speed in film emulsions or lenses to get any moving images - and it's the constant movement that adds so much to the magic of this remarkable display.

Dominic Case



Tony Coldwell wrote :

>I'm shooting in Churchill in Canada in November and hope to capture >the aurora borealis on film. I'm taking my Aaton and an intervalometer.

This may be totally useless, but it's worth a thought - You need to record something a long way away that gives off very little light. Ideally (I guess) you'd want to be able to record it in real time (has this ever been done satisfactorily before?).

I've seen the Northern Lights from an airliner flying over Northern Canada, and the motion of them was the most fascinating thing. If you use either very fast film stocks, or highly sensitive video equipment, even with the fastest lenses you can get hold of, your main problem could still be noise - that is, either film grain (fast stock/push processing) or electronic noise.

From my experience trying to get rid of obvious static grain from still images that are to be used for visual effects plates (matte paintings etc), I find the best method to get rid of grain on a static subject is to average together several frames. This is achieved by mixing together say 12 frames so that the resulting frame consists of an even mixture of all 12. The resulting totally grain less image really is something to behold - particularly a 6k VistaVision frame!

So - how does this help us with the aurora? Well - my barmy idea is this - set up several identically specified cameras (film or video), same lenses etc, and record the action in the sky, boosting the sensitivity of the medium as much as you can while not getting too worried about the very extreme grain/noise you will get. Then take all the "parallel" material into digital post (preferably someone who's thought about/done this stuff before), and average all the cameras together. Hey presto - some aurora footage as never seen before. Think of it as shooting via 10 lenses all at once, with all the light gathering power that would give you. This only works because the aurora effects are distant enough for parallax not to be a problem. And it only works if you can rustle up enough cameras/film - that's the real trick, I suppose!

Paddy Eason
Double Negative, London



Hi Tony,

Well, the first piece of advice would be to rent "Picture of Light" (director - Peter Mettler) since there's a lot of info as well as beautiful images and a good film.

As you may have guessed, bring plenty of batteries, maybe one with more amp hours than an np-1, this way, you won't have the added hassle of changing the batteries in the middle of the night. Even though you're using next to no power, the cold will seriously drain your battery.

If your body is not an XTR prod, you may want to get the camera winterised (or get it winterised anyway), which I believe means using a thinner camera oil. All in all, when shooting time-lapse, there is not as much of a worry since the mechanism is moving much slower.

I'm not sure that a fluid head would work all that well, especially in sub-zero temperatures over a long period of time, so I would advise on getting a mini-Worrall or the like. Besides, you're not going anywhere. You would need a somewhat functioning head since you may have to reframe slight as the aurora moves around.

I cannot emphasize on how warm your clothing should be, since it will not only be cold, but you would not be moving around much, and this will be over a period of hours.

Common sensical advice - keep away from lamp posts, etc.

Keep in mind that nature is VERY active in that area. Bears are commonly spotted in that area. If you have good camping skills, then you should be all right.

In terms of exposure - well, it depends on your stock and how long you can keep that shutter open. What kind of lenses will you be bringing with you?

It would be a good idea to keep that internal light meter switched off, since you may very well get fogging with that speed.

Stay neutral while shooting, keep your thoughts together, bring a book. Listening to the radio helps, especially if you're a short wave fan. Otherwise, plenty of tapes, CDs, what have you.

Patience.

It would also benefit tremendously to work with a partner, in case anything happens, from the mundane to the serious. I shot star fields at night over a period of 10 months, and caught the Aurora Borealis on two occasions. Mind you, I was shooting around Southern Quebec.

Let me know if you have more specific questions, I'd be glad to help.

Regards,

Duraid Munajim
Montreal, Canada



Paddy suggests :

>So - how does this help us with the aurora? Well (my barmy idea is this) >set up several identically specified cameras (film or video), same >lenses etc, and record the action in the sky

Ok, assuming he's not the only crazy one, I would ask whether the motion of the aurora is slow enough that if you shot two or three sequential frames for each frame that you wanted (if this were to be an intervelometer sort of shoot) would they be close enough in time with respect to the motion to be able to "frame average" them for grain reduction?

This would only work if :

1/. Motion is very slow (never seen an aurora that didn't have tires)

2/. Intended footage was to be intervelometer as opposed to low frame rate continuous in the first - place.

A geared head seems like a good idea - if you set the wheels on the lowest gearing and you find you do have to re-frame, you might be able to get a usable pan by very slightly incrementing the pan wheel between exposures until you get to your new position.

If you mark divisions around the handle, or cut a cardboard disk to put behind the handle with divisions and use a bit of coat hanger to make a pointer, you might actually get bearably usable pans - poor man's motion control.

Mark "crazy too" Weingartner



Paddy Eason wrote :

>This may be totally useless...average together several frames...set up >several identically specified cameras (film or video),

>same lenses etc,


You can rent an off-the-shelf mirrored dual camera system such as Clairmont's device. Or zero-out one of the many systems that are made for stereo (3-D) vision and you have (sort of) doubled your resolution right there.

Additionally, there many types of custom multi camera rigs out there built for special venue surround theatres that have many projectors/screens. Those systems use "heads" that are designed to simultaneously shoot with camera counts like six, nine, or twelve. Some even shoot into prisms to approximate a coplanar (dare I say nodal) point.

Again, presto!

Oh yes, don't forget the Motion Control for time-lapse camera

movement.

Eric



So here's what I know about the aurora borealis film that was shot in Churchill. (I have found it titled "Pictures of Light" as well as "Picture of Light")

It was made in 1994 by Petter Mettler, a DP/Director that shot Atom Egoyan's first couple of films. I haven't been able to find a source for the film on video, but I would bet it is available somewhere in Canada (Mettler is Canadian). Below are a few links that will show you a few stills and might give you some clues to help you find this doc :

http://www.checkout.com/movies/title/info/0,7695,873440,00.html

http://members.cruzio.com/~akreyche/tchkpm.html

We live in a time where things do not seem to exist unless they have been captured in an image. Peter Mettler commentary from Picture Of Light

Good luck,
Rob Barocci
DP-New York City



>I would ask whether the motion of the aurora is slow enough that if you >shot two or three sequential frames

It's many many years since I saw or photographed an aurora. I guess they haven't changed much though.

My recollection was that I was using exposure times of a few seconds, on ASA200 reversal film (the fastest you could get then) with a not very fast lens (around f2.8??). Today, boosting the stock to 1600 (800 pushed a stop), and with a 1.4 lens, you might get down to 1/6th second or so.

You might even be able to get an exposure meter to register the brightness of the aurora! I made do with bracketed exposures and threw away a lot of frames.

I remember describing the movements as similar to a bank of searchlights scanning the sky in random patterns. You'd need several frames per second to reproduce it with any grace.

So Mark's idea isn't out of the question, but it sounds marginal. It may be that you'd need Paddy's idea of several cameras to get enough frames per second - or maybe he could arrange some motion interpolation between frames.

Dominic Case
Group Technology & Services Manager
Atlab Australia
web: www.atlab.com.au



Mark Weingartner wrote :

>Ok, assuming he's not the only crazy one, I would ask whether the >motion of the aurora is slow enough that if you shot two or three >sequential frames for each frame that you wanted .....

I think at the end of the day, that doesn't give you any more than by simply shooting at a slower frame rate:

eg - shooting 24 fps, then in post averaging each alternate batch of 12 frames (=shutter angle of 180 degrees) to produce material that is sped up to playback 24x fast...is equivalent to...- simply shooting at 1 fps, and exposing each frame with 12x more light.

The end result would probably be similar, but the latter is a heck of a lot easier, and no need for post work.

I think with my silly multiple camera setup you would get the advantage of many pieces of film, and many lenses all gathering information to be averaged for noise reduction later.

I think the idea about using multiple stills cameras (with magazines) could be a good one, though a fair amount of post work would be involved in simply lining up the individual frames (not a proper pin registered film transport).

Paddy Eason
Double Negative, London


Sponsored by :-

 

 

 

 

 

 

 

CML Home CML-Tests Home

© copyright CML - Cinematography Mailing List all rights reserved