WATER IS white. “Thanks, Captain Obvious,” you are probably thinking. 🙂 Although it does seem like a pretty obvious statement, many photographs “suggest” otherwise. Often a color cast is subtle, but it is there. In addition to being white (well, technically clear), water is a reflective medium. We often want our water images to reflect colors and shapes from the surroundings. Good when it is part of the planned composition. A lot of water images, though, have an unnatural color cast that really needs to be corrected. There are a number of ways to fix them. Generally, they all involve adjustments to saturation, contrast, and brightness. The most effective way I know to remove a color cast from an image area that is supposed to be white is to de-saturate it. Sometimes, just that adjustment suffices. The trick, though, is to isolate just those areas of the image you want to de-saturate. Otherwise, de-saturation can take some of the “life” out of the image. Increasing contrast will often restore (at least some of) that life. You may also find that you need to reduce or increase image brightness. As culinary recipes frequently advise, adjust brightness (up or down) “to taste.” In order to make these adjustments “targeted” to just the areas that should be white, you are almost always going to need to use a selection/mask process of some sort. Options include making selections (and probably feathering them appropriately), using Photoshop’s Adjustment Layers and then painting on the mask, and using a luminosity mask, among others. It is easy and effective to make targeted adjustments using Adjustment Layers, by creating a layer (e.g., a hue/saturation layer) and making global adjustments to obtain the “look” you are seeking for the water. You simply ignore the changes (which could be garish) to the other parts of the image. You will “fix” them in the next step. Once you have the water as you like it, click onto the mask that was created for your adjustment layer. Control-i (for Windows) will invert the mask, hiding all of the changes you just made. You then can use a soft brush with low opacity to reveal the areas you want to show being changed (you could skip the invert step and use a black brush to hide the areas you want to go back to “normal,” but I prefer – in this instance – to hide all my changes and brush in (“reveal”) the changes where I want them. Here is a tip it took me a long time to learn to use to my advantage: use the opacity slider on the layer in the panel to reduce opacity while you work, so you can see the edges of the area better, and the effect of your changes as they blend with the original layer. This technique works well in most cases. In some cases, as you will see below, you may have better success making a selection before creating the adjustment layer. This will now only include the selected area in the layer mask (and the selection will only be in the mask – if you want to do the same selection again, go to the Select drop down in the top menu and choose “reselect”).
ALTERNATIVELY, YOU can use an “app” that has been created for image adjustment (I like and use the NIK collection, by DxO, but there are others that will also work well, with varying amounts of “automatic adjustments” vs. user-controlled adjustments). This is the option I choose most of the time. It is easy, and the selection/masking process within the apps are better than I can usually do myself. My own “go-to” tool for correcting colors in water, fog and clouds is a plugin app for photoshop called NIK Viveza2. Originally offered by a company named (aptly enough) NIK Software, it was purchased by Google in 2012. The software was actually a “collection” of a few apps, including Viveza2, a filter program called ColorEFx, a B&W conversion app (Silver EFx), and a noise reduction app (“Dfine”). I mostly use Viveza2. Prior to the Google acquisition, there was a cost to purchase and at first, I pooh-poohed it when several of my friends were extolling its virtues. I had by then climbed a fair way up the Photoshop learning curve and using other techniques and actions, I felt that I could get the results I needed without resorting to what I though was “gimmicky” software at the time (but for those who are counting, I was wrong). 🙂 Once I tried it, I never looked back. Another one of the better options available to “power users” were something called “Luminosity Masks.” They basically were some sophisticated selections used to “mask” just the areas you wanted a photoshop adjustment to affect. Unfortunately, I never really got beyond some very simplistic uses of the masks. I know they are useful. My friend, Kerry Leibowitz uses them extremely effectively. But something along the way goaded me into the “free trial” of the NIK software. I was immediately convinced by its ease of use and general effectiveness and bought it. After the Google acquisition, Google made is a free download for a time. By 2018, Google had discontinued it, and DxO Software acquired the rights and released their “new and improved” version. My original software still works, and I have not felt the need to purchase the newer software – yet. It seems like each time Photoshop releases a full new version, getting the plugin in the right place and working properly becomes more complicated. There may come a day when I have to purchase the newest version. But for now, I use it often. There are competitive programs that will also work on these issues. For the most part, I haven’t been nearly as impressed with them mostly because they are not as “targeted,” and because they really are essentially filters (albeit with some pretty useful adjustment and customizing capability). But I think the Viveza2 is a stronger, more flexible and just overall better program when it comes to these “apps.”
THE FIRST image is Scott Falls, in Michigan’s Upper Peninsula, near Munising. It is easy to see from the side-by-side presentation that the waterfall has a pretty distinct, mostly blue (but I also detect some unwanted magenta) cast. When viewing it alone, it is not so obvious. When “curating” images for post-processing, it is important to consider color casts and try to visualize the true color of the flowing water – as your eyes saw it, rather than as the sensor recorded it. I used Viveza2 to “fix” the blue cast. It is really simple to do. Just open the Viveza2 (in Photoshop, it is in the Filters menu – but it can also be used as a stand-alone and plugs into a few of the other popular post-processing software). Viveza2 uses a feature they call “U-point” technology, which creates masks using essentially AI. You simply click on the little circular target (“control point”), and then place it (and click) in the image, where you want to make an adjustment. This creates a drop-down menu within the image. The same adjustments can be made in the panel on the left, as well as hiding and/or deleting the control point, and linking multiple control points to make group adjustments. The in-image drop down includes several sliders one of which allows you to adjust the size of the area affected, as well. It is pretty smart, selecting the color or range of colors you want, even when there are other colors present. It takes some playing around to get the feel for it. I place the target in the area(s) of the water that needs color correction, and then use three of the drop-down menu sliders: Saturation, Contrast, and Brightness. It is mostly a function of saturation and I start by desaturating the image, often as much as 90% or more. You will see the water turn white. You may also see other adjacent areas desaturate. Use the size slider to control this. I often have a series of many very small adjustment points in a waterfall. The other two adjustments are very much “to taste.” I often add contrast (10-15% at the most). Sometimes I add brightness (again in very small amounts). It is a technique that works very well. When you use Viveza2, it automatically creates its own layer, so you can also use the opacity slider on the layer, and/or paint on the layer mask to make further fine-tuning adjustments.
THE MAD River image has lots of problems. I chose it as an illustration, but probably would not choose it as an image to display on my website. The thing that makes it instructive here, is that it is a pretty “busy” image, with lots of individual areas of off-color water to be targeted. Using my Viveza2 plugin here would not only be “fiddly,” but would probably ultimately not come out the way I wished. Here, I used one of Photoshop’s selection methods called “Selective Color.” In the “Selection” drop down menu, I chose “Color Range,” and then clicked the eyedropper (the one on the left will select the range where you place it in the image) on an area of the blue-gray water that I wanted to color correct. This may take some trial and error. Adjusting the “fuzziness” slider in the dialog box will affect the range of colors around the selected color. More fuzziness will create more lattitude. Also, you may want to go to the tools menu (on the left in my illustration) and click on the eyedropper tool. Then, up on the options bar on top, change the sampling to 3×3 sample. Once I made the color range selection, I made a saturation adjustment layer and reduced the saturation. I then made a contrast layer and increased the contrast a bit. This resulted in the water change I wanted, but it also desaturated the surrounding rocks and dropped leaves more than I wanted to (remember that the adjustment layers will only consist of the selected area, but this shows that perhaps my settings were a bit “too fuzzy”). Using the same selection as before, I inverted the selection (Selection/Reselect/Inverse) and created a second saturation layer, in which I adjusted the saturation back to taste. The result created too much saturation in the green leaves in the right foreground. Using a brush on the layer mask for the saturation layer just created, I brushed out (black hides) the areas I wanted to tone down. I usually use a reduced-opacity (between 30-50%), very soft brush for this. Another way would have been to bring the file back into Viveza2 and select and desaturate the oversaturated elements. Fiddly. You probably should make a determination if the image is worth the work before going too far down the road with it.
I USED Viveza2 again on the Glade Creek Gristmill. Not only was it convenient to use that software to correct the water, but as you can see, I was also able to take the slight blue cast out of the weathered wood building (placing additional “Control Points”) and add some contrast and brightness to the surrounding foliage. Our trip to Babcock State Park in West Virginia (site of the gristmill) was mostly a dreary, rainy experience. The gray skies contribute to the shadows and consequent blue cast of the water.
FOG IS another tricky situation. Essentially the same as clouds, we generally associate it as white (but perhaps subtly less bright than clouds). The same adjustments will be used here: saturation, contrast, and brightness. For the most part, I use the same tools and techniques with fog as I do with white-water. Because the water is more vapor, and often less reflective, care must be taken not to overdue he adjustments here. A favorite example for me is my shot of the barn and mountains scene in Barton, Vermont in 2010. We had a wonderful fall morning for photography, with the early light producing saturated colors, and the cold temperatures following a very wet period of several days, creating spectacular fog (or perhaps low-hanging clouds). It was a couple years after I made the image and posted it here and on my LightCentricPhotography Website that I actually became aware of the problems with the image. With this image, it was a pretty simple fix. Just open it in Viveza2, set a few control points across the cloud bank, desaturate, add a small amount of contrast and brightness. Viola’! A much better looking image, I think.
THERE ARE more subtle fog examples. My Obow Bend shot of the Snake River in Jackson, Wyoming is perhaps my best example of very subtle fog. When shooting around water in the early mornings, we often hope for a cold night and a cool, clear dawn. If the conditions are right and the water is warmer than the air, conditions often produce really nice, surface fog. Conditions were there this morning. The original image – perhaps because of the shadows – rendered this fog what I thought was an unnatural bluish color. Viveza2 did a great job of removing the blue cast, while preserving the reflective colors in the water. At first glance, the viewer may not even realize the fog is there. It is. 🙂
THE VERY small, shallow ponds (they are called lakes on the map) found in the relatively flat Hiawatha National Forest in Michigan’s U.P., create similar conditions, particularly in a cold morning following a clear night. I have been to, and photographed Moccasin Lake in the forest a number of times. The image here though, is one of my favorites and came during a very cold sunrise. The orange/pink color of the early rising sun colored the fog. In the original image, the camera sensor recorded what I felt was too much color information – enough so that it looked unreal to me and was not what my eye/brain computer “remembered.” Again, Viveza2 to the rescue. But in this case, I wanted some of the color to remain, in keeping with the warm morning sunlight in spite of a very cold morning. For that reason, the saturation adjustment was gentler, leaving, I hoped, a more subtle color correction. Wanting the foliage colors to show a bit through the fog, I made a stronger correction on the surface fog than I did on the low-hanging fog.
AT THIS point, a reader might think I believe that every image containing clouds, whitewater, or fog has to be “fixed.” Not so. There are certainly images in which the water-based medium takes on surrounding colors, and we want them. There are certainly times when reflected color will add drama and atmosphere to the image. In 2010, I photographed a small, roadside pond just south of Barton, Vermont on U.S. Route 5, named “Bean Pond.” Again, I have shot here several times and come away with what I think are some very nice, atmospheric images. As the shot below illustrates, this is an instance when rendering the water vapor pure (or near pure) white would not be “fixing” the image. It would, in my view, be deleterious. I did do some work on the fog hanging over the water surface. But the morning-lit cloud is what I think gives the image its essential character.
LARGER EXPANSES of water may well require a wholly different approach. The color of the water surface of a large body of water is almost wholly a function of the reflection of the sky. That is why water often appears to us to be blue. If the sky is dark, cloudy or overcast, water will turn different shades of gray, green, brown or even black. In the rather dramatic sunset over St. Kitts in the Caribbean, the water surface has a mixed gold and magenta cast. Often, we try to remove magenta color casts from images. But to do so here would really defeat the spirit of this image.
ONE LAST water-related subject remains: Snow. Ironically, I lived in Michigan and Vermont for most of my adult life, but never really got many snow images, so my winter portfolio is weak. I do have a couple. I have a friend who is a talented photographer. He comes from the film days, when you kind of took what the film gave you. In the past several months, he has posted a few very nicely composed snow scenes. But – perhaps because I am sensitive to it – I immediately noticed a very strong color cast on some of them. The most common casts I see in snow scenes are blue (generally produced by shadows and low light conditions) and red/magenta (which is a factor of the kind of light that often shows up at the “golden” hours of early and late day. If you are cognizant, and have some context, the casts can be quite jarring. The same principles as discussed above will stand you in good stead when color-correcting snow images. Carefully desaturating the snowy portions of the image, with some subtle contrast and brightness adjustments will usually do the trick. Be careful not to overdo it and make the snow seem dirty. Also, be aware of white buildings, and very contrasty, bright colored buildings. You may need to do some additional work to make them blend in properly. But it will be worth the effort.
COLOR CASTS in snow – especially blue – are tricky. Again, the eye fools us. Our eye/brain connection is the most advanced processing engine that exists. It takes our thousands of memories, along with our innate knowledge of natural colors, and tells our eye that the snow we are looking at is white. Digital sensors don’t do that (unless we tell them to – see, for example, my “white balance” discussion below). When an image is recorded in its “raw” pixel format, it captures the colors it sees. In the Point Iroquis Lighthouse image, the dominant color in the photograph is blue. Ironically, probably 75 percent of the frame is (should be) white. The relatively low-angled, pure-blue, morning sky creates deep shadows, and those shadows pick of the predominant blue color in the image (interestingly the off-white color of the lighthouse is well enough lit that it pretty much looks white. But the snow? Looking at the comparison shot, its blue. Not much doubt when we have context.
SOME MIGHT say I am missing the proverbial boat, here and that all of this is a matter of getting the correct white balance. They wouldn’t be wrong, but I never do that in-camera. You definitely need to pay attention to white balance. To me, that is for shooting in jpeg, tiff or some other “pre-cooked” file type. I never do. I always shoot in raw, and in raw, I can (and sometimes – but rarely – do) correct white balance type issues in post-processing. The next time you photograph a scene that has water, fog, clouds, or even snow, think about whether the color you see on your display is truly the appropriate color for water and water-related images. Don’t be afraid to play around a bit with the saturation and contrast, as – without comparative context – our eyes can fool us.