Color is a perception, not a physical thing, so It is unsurprising that our human brain and the camera's "brain" don't always agree about color.
Was that the "real" color? It's a good question, but the answer is a bit complicated. The color blue is uncommon in nature. Certain flowers, yes. Skies, of course. But rocks?
It is not rare in the Escalante canyons to see a deep, almost cobalt blue color on canyon walls that have a lot of desert varnish. Desert varnish is a dark, shiny coating of minerals that have slowly built up over centuries. The minerals are brought to the wall by surface water, so desert varnish is usually thickest and darkest where water pours off the cliffs.
Desert varnish is reflective, so in those situations where the sky is blue and the wall's angle is right, the blue sky will be reflected to the viewer and parts of the wall will appear blue.. This phenomenon is pretty common in the Escalante canyons, if you look for it. What is uncommon in this photograph is the amount of blue. This wall is exceptional in the amount and uniformity of desert varnish combined with its angle to the blue sky.
In fact, if you like blue walls, this one is about the best I've seen in the Escalante canyons—which is, of course, the reason I photographed it.
Hyde's Wall, East Moody Canyon. Photo: © Donald J. Rommes
People who see this photograph and who are unfamiliar with desert varnish have a hard time believing the color is real. That's understandable. The photographer may have intentionally colored the blue for artistic reasons—something that is easy to do in the era of Photoshop®. That's the prerogative of the artist.
However, if people were made to trust the canyon wall is truly blue, their appreciation for the light in these marvelous canyons may be enhanced. That is what I am after.
By getting the color temperature correct, familiar objects like trees and bushes will have the "right" color which may help a skeptical viewer accept the more unusual colors—of blue walls, for example.
I first took a photograph (later published) of this wall with a view camera using Fuji Astia transparency film. The wall was indeed blue, but the measured color temperature (taken with a color meter) was very warm. I used a color balancing filter to "correct" the light somewhat and help the slide film record color as I perceived it in the field. Even so, it still had a slight reddish color cast.
I returned to this spot recently—this time with a digital camera. It was late in the day and, admittedly, I was hurrying to get to camp in another canyon before dark. My digital camera was still set to DAYLIGHT color balance. I exposed three side-by-side images with the intent of making a panorama once back at my computer. When I opened the file in Lightroom, with no adjustments, this is the left-most of the three scenes:
Screenshot from Lightroom shows this was exposed with a white balance set to 4700K and no that other adjustments were made to the image.
The wall is indeed blueish, but the whole image is far too red—much redder than I remembered the place being when I was there. That's because the light illuminating the scene had a very low color temperature. The source of the color was the warm light of late afternoon bouncing off red sandstone walls on the opposite side of the canyon.
By setting my camera's white balance to DAYLIGHT, I was asking it to show me how the scene would look if illuminated by sunlight (approx. 5000K) and not the reflected light .
The light at the time had a very low color temperature (2700ºK) for the reasons given above. Light of that temperature is usually perceived by us as red. Yet, when I was there, I could not tell the ambient light was red. The scene looked like the uppermost photo in this blog—blue wall, yellowish rocks.
That's normal though.
Color is a perception. It is the result of the brain's subconscious interpretation of signals generated by light receptors in the eye. We do not perceive the color of something simply by the color temperature of the light that illuminates it. if color perception was based solely on color temperature of the light, we would see a strong red cast in East Moody and in Hyde's Wall.
But we don't.
In person, what I saw (perceived) closely resembled the first photo with the blue wall.
In humans, color is determined by comparing the relative strength of the signals coming from the three color receptors n our eyes—much as a digital camera does.
That's a potential problem for photographers. The light illuminating the wall and the talus slope truly had a low color temperature. If we recorded the scene with daylight film, or a digital camera with the white balance set to daylight, the resulting image would have strong red cast—even though you wouldn't have noticed the red cast in person.
Film with a red cast is not easy to correct in printing. The color of a RAW file can be easily and non-destructively modified with software, but what should the "real" color be? The photographer would have to rely on memory to get the color close in post processing. Better to recognize what is going on in the camera and try to get the color close than guessing later—particularly with unusual color in a scene. .
That's because people are very familiar with certain colors in the landscape—the color of grass, for example, spring leaves, blue sky—and tend to reject photographs that get them wrong. Many of the colors in the Escalante canyons—especially with all the reflected light from intensely colored canyon walls—are outside the experience of most people and are difficult for them to accept as real. If you want the viewer to accept unusual colors as real—blue stone walls, for example—then you have to get the colors of the common things in the scene right.
In this particular case, changing the color balance of the RAW image in Photoshop® from about 4700K to 2700LK got the colors "right." That determination was made by getting the greens right. Since I made no other changes to affect color, if the greens were accurate, slo was the color of the blue wall.
Here's the adjusted RAW file using 2700K as the white balance:
Screenshot from Lightroom after adjusting the white balance in the previous iteration from about 4700K to 2700K. The greens look "right" so the blue must be the color of the wall—which, after all, is what I perceived at the time. Photo © Donald J. Rommes
I synched the adjustment from this file to the other two files in the panorama and stitched them together, giving this result:
A few more little tweaks, and we get the image at the beginning of this post.
So, the "real " color? Well, the first image is the color I perceived when I was there in person. Since color is a perception, I'd call that "Real."
The photo with the strong red cast is a good representation of the physical property (color temperature) of the reflected light in the canyon at the time. But it doesn't replicate the human experience, the human perception, so it's not "real color."
Comments