It may surprise people how many factors need to be considered when making an image of the night sky. What did you have to do?
Yeah, you have factors like when the object you're going to photograph is going to be up, where's it going to be in the sky, when the sky is going to be dark. For the picture of the Owachomo Bridge in Utah I went out and got some very nice astronomy software that told me when the Milky Way was going to rise in the location where I was shooting. When I was there in March, the southern Milky Way rose at 2:30 in the morning, and that would give me two and a half to three hours before sunrise. If you are doing dark-sky stuff like the Milky Way, you absolutely cannot have any moon. Any moon will turn the sky blue. Not only was I looking at the weather conditions, but the moon cycles and when the moonrise would be versus when the southern Milky Way would rise.
What were the technical challenges?
For the Owachomo Bridge I wanted both the landscape and the stars sharp. It has only been in the last couple of years—and in some cases only the last year—that there have been cameras that could do this. In the past when you did pictures like this you'd do a much longer time exposure and put the camera on a clock drive, then photograph just the Milky Way. If you did a five-minute exposure to get the Milky Way sharp, anything on the ground would blur. And I didn't want that—I wanted the landscape and the stars both sharp.
What kind of camera did you use?
I used a Nikon D3. It’s the current champion of high ISO/low noise. They also had a 14mm lens—which means really, really wide-angle—that doesn’t have very much falloff in illumination at the corners. Also it could produce sharp images of stars out to the corners. Usually wide-angle lenses weren’t very good out at the corners, so you would turn it down to f/8. But I can’t turn it down to f/8, because that would make the exposure too long and everything would blur. This 14mm is an f/2.8 lens, and good at f/2.8—a great combination. A 14mm lens also meant that I could do a longer exposure, because the wider the lens, the longer the exposure you can do before the stars are visibly blurred by the Earth's rotation, leaving little streaks. With a 35mm lens you're pretty much limited to 30 seconds; with a 14mm lens you can probably do a minute, maybe even 90 seconds before it starts blurring badly. That's how I got all the scene sharp. And this long exposure, which I think is 90 seconds, gave me time to paint the arch with a flashlight, lighting it up.
How about the picture of Chicago from above—how challenging was that to make?
I was hoping for a picture of a city where I could see the way the city was lighting up the clouds from below. It looks bright, and it is bright to your eye. But to the camera, even to the high ISO, the exposure on this was still something like an eighth or a tenth of a second at f/2, and I had to go out and buy another lens that specifically had a fast enough f-stop to do that. You have to understand that an eighth of a second is quite long—most people wouldn’t take a picture at an eighth of a second handheld, even if they were standing on solid ground. I’m in a small plane and doing an eighth of a second, which is supposed to be impossible. Usually from small planes, people say you don’t want to go below 500th of a second to get things sharp. So this was way, way out on the edge of what anybody would think would be doable.
Would you say the images made for this assignment are good examples of digital photography's potential?
Yes, you couldn’t do this on film. Years ago I tried doing the Milky Way on the fastest film that was available at the time and failed. The fact that film gets less sensitive during long exposures just kills you. When I was up photographing Chicago, it had this beautiful sparkly glow to it. But if the exposure isn't right, a picture of it doesn’t look like that at all. The great thing was to do the exposure and see what it looked like. The look of the picture changes drastically with the exposure. Up there that night there weren’t places where I could meter off of something very well. So I was very pleased with being able to see the image and get the feedback of what it looked like, whether it had the visual and emotional impact that it had to my eye.