Although it was more than 25 years ago, it really doesn’t seem so long since I made this Fujichrome Velvia slide of Conrail’s BUOI (Road freight from Buffalo to Oak Island) along the former Erie Railroad in the Canisteo Valley.
I’d followed the train east from Rock Glen, New York. Steady snow made for slippery road conditions so I took it easy.
Here I’d caught up with the train, which had reached the newly created siding east of Adrian, that would soon become ‘CP Adrian’ (CP for dispatcher Control Point).
Work was under way at the time, but the new color light signals hadn’t been commissioned and the old semaphores that had governed movements under rule 241 (current of traffic) remained in place, but deactivated.
Working with my Nikon F3T and 105mm lens, I exposed this view as the train waited for permission to proceed east.
Velvia was a finicky film and it was tough to nail the exposure in some conditions Getting the snow exposure right was tricky, but since the train wasn’t moving I made a bracket—in other words I exposed several slides with slight exposure variations. You can see that it was relatively dark by the illumination in the number boards on 6118.
I thought in the heat of high summer, it might be a refreshing time to present some frosty views from last January.
Previously, I’d posted some black & white photos exposed during a lake effect snow squall at Brookfield, Wisconsin on CP Rail’s former Milwaukee Road main line that I’d made on a photographic adventure with Trains Magazine’s Brian Schmidt.
That same morning, I’d also worked in color using my FujiFilm XT1.
Here’s a view of a westward empty grain train passing the old Brookfield station led by brightly painted BNSF GE’s.
It’s a stark contrast from the leafy trees and high summer temperatures of today.
I’m not talking about stripping down and running naked through the snow. That sounds like a recipe for frostbite, hypothermia or worse! Rather, I’m focused on how to best select exposure when working in winter situations. Snow is especially difficult to work with for several reasons. First, it’s abnormally bright and results in high contrast situations that is both difficult on the eyes and the camera sensor. Second, most camera meters aren’t designed to work with fields of white, so tend to recommend the wrong settings. Third, for many photographers, making images in snow is an infrequent experience, and one that tends to lead to uncertainty and higher rates of exposure error.
My approach to snow photography stems from years of practice. In general, I take the information provided by camera meters as advisory. I rarely rely on automatic settings without some manual adjustment. Why? I’ve learned to carefully gauge exposure and apply settings manually. Furthermore, I’m distrustful of automatic metering, especially for railway photography, because the automation is programmed to deliver adequate imagery other than what I’m trying to achieve. Perhaps no other situation is as difficult for a common-meter to gauge as sunlit snow imaging.
Many years ago, my father lent me a Weston Master III, and instructed me to wander around the house making exposures and write them down. No photos were exposed. I was about nine and I found this exercise confusing and frustrating because I didn’t understand what I was supposed to be doing. However, I overcame frustration and learned to use the light meter. A decade later, I had the opportunity to learn Cibachrome printing (used to make vivid prints from color slides). At the time, I was primarily working with Kodachrome 25, which I’d been taught to nominally underexpose to produce more saturated colors.
Translating Kodachrome to Cibachrome was revealing; I’d found that my rich, slightly-underexposed slides, which when projected on a nice bright screen looked fantastic, were in fact rather difficult to print. The biggest issue was contrast. While under-exposure may have enhanced the color saturation, it also made the image more contrasty. So while it turned out that my old theory on underexposure had it flaws, I discovered that slightly overexposed slides printed very well. I needed to determine ideal exposures in order to make optimal slides.
Aiding my efforts was my notebook; I’d been recording my exposures for years, but with the Ciba exercise I began making even more detailed notes, recording slide exposures to the third of a stop. Eventually, I assembled a chart with ideal exposures for Kodachrome 25 in various lighting situations. In general, I’d discovered that to make prints, slides needed to be about 1/3 stop brighter than I’d been making them for projection. All very well, but what does this have to do with making digital images in the snow?
Exposing Kodachrome is history, but the lessons I learned from this material still apply. (The short answer to the question was that snow in bright daylight should be exposed at approximately 1 ½ stops down from the full daylight setting without snow; thus with Kodachrome 25, if my normal daylight setting was f4.5 1/250, my snow exposure was about f8 1/250 +/- 1/3 stop). Many of my slides have appeared in books, magazines, as well as here on Tracking the Light. Take a look at my recent book North American Locomotives for some top-notch printed reproductions of Kodachrome.
Digital photography offers some great advantages over Kodachrome, including the ability to review images on-site—thus removing the uncertainty of exposing slides and having to wait for days (or weeks) to see if your exposures were correct. It’s now easier than ever to make good snow exposures and learn immediately from miscalculation. Related to this is the ability to use a digital camera’s histogram as an on-site exposure tool.
Histogram? Yes! This is perhaps the greatest feature on my digital cameras. It allows me to set my exposure ideally, allowing key images to capture the greatest amount information, thus minimizing detail lost through unwanted under-or over-exposure. Positively invaluable when making images in the snow.
Today, before a train enters the scene, I’ll make a series of test exposures and judge them by the output of the histogram. This allows me to refine my exposure to a point that exceeds what I could have achieved with my detailed chart and Kodachrome. In my next post, I’ll detail this process with more examples.