top of page
Search

Drift Feeding in Salmonids and the Concept of "The Window"

Although it is too simple to look at salmonids as strictly drift feeding fishes - meaning that they hold position in a stream waiting for the current to bring aquatic invertebrates into their feeding window - it is a very useful way to begin to think about why salmonids are where they are. The concept is simple, drift feeding fishes hold in a position that minimizes energy expenditure, yet maximizes energy intake. This is the basis of net energy intake (NEI) models - an optimal foraging theory model. Optimal foraging theory essentially states that individuals maximize their net energy intake - that is they spend as little energy as they can to gain as much food - as many calories - as they can.

As trout anglers, we are all a bit familiar with trout and how they feed, or at least we better be if we want to be successful trout anglers. There are any number of things that help anglers be successful - or keep them from being successful. Certainly one of those is knowing where trout hold and getting your flies to where the trout are. Then the goal is to present an imitation of a food item in a way that does not alarm a fish. I think trout are looking for negative triggers at least as much as they are looking for reasons to eat a prey item in their window.


Defining "The Window"

Fish swim past an underwater window, surrounded by plants and branches. The scene has a serene, aquatic ambiance with soft green hues.

The "window" can be defined as the three-dimensional space - a volume of water - in which prey is both visible and a fish is able to capture that prey item. The size and shape of this window is highly dependent upon the fish's environment. Most notably, it is a function of water velocity as well as the fish's ability to move to capture prey. The size of the window generally increases with fish size - though larger fish are generally more selective within that window.

Diagram of a fish with labels: Drift-Feeding Foraging Window, Current Velocity, Water Clarity. Shows effect of water conditions on vision.

While current velocity is the main factor that affects the size of the window, it is not the only factor that influences the volume of water within the window. Most obviously, water depth will influence the volume of water in which a trout is able to forage. In deeper water, they may not even be able to see your dry fly on the surface. Another important factor is water clarity. In turbid waters, trout have a smaller foraging window and as a very good friend and graduate school colleague showed, they often expend more energy actively searching for food in turbid conditions rather than holding a drift-feeding position (Sweka and Hartman 2001a, b). Similarly, you may have noticed at dusk and dawn, trout are often holding closer to the surface when aquatic insects are hatching or returning to the stream to deposit their eggs. Part of this is that their foraging windows are smaller at lower light intensities. Another part is that the trouts' predators can not see them as well under low light conditions so this position, once unsafe, is now safe for them.


Foraging within the Window


Put yourself in a drift-feeding trout's world for a second. You are living in a world in which you are well suited - we move nowhere near as effectively through water. However, due to the constant downstream movement of water, staying in place requires a constant energy expenditure. As you sit in place - the smart trout holds in a place that requires relatively little energy expenditure to hold position - a conveyor belt of food is moving past you. However, on this conveyor belt is not only food but debris like pine needles, pieces of aquatic vegetation, and other non-food items. Each time you mistake something that is not food for food, you are expending valuable energy for no reward. Do this too often and you do not survive.

Hands hold a pan with water, small stones, and debris - these are the stomach contents of a Chinook Salmon. The background shows buckets on a rocky surface.
Stomach contents of a Chinook Salmon - notice that not all of it is food. NOAA image from the Creative Commons.

Having assisted friends that conducted trout diet surveys, I know that not all that trout eat is food. They fairly regularly capture debris and other non-food items. As fly anglers, I suppose we should be happy that they are not always able to discern food from our imitations of food - many of which do not look much like mayfly, caddis, or other aquatic prey items. I will not pretend to know what trout are thinking, but I will venture a semi-educated guess that trout are probably looking for both evidence that an item is food and a lack of there being "something wrong". In slow water, trout have time to more thoroughly examine potential food, whereas in riffles, trout have less time to inspect each potential morsel. This is why the fish-counting competitive anglers have largely tailored their methods to fast moving water. Fishes are there to forage - they are expending too much energy not to be feeding - and they have to make quick decisions. "Euronymphs" are meant to get to the bottom quickly and look enough like food that they get eaten - there is nothing "magical" about it.


Drift-Feeding Models


I entered graduate school just after / during a rebirth of drift feeding models. Scientists like Kurt Fausch and Nick Hughes certainly were not the first to consider drift-feeding models, but they certainly helped make major advancements. Fausch (1984) put out the idea of net energy intake and was able to test it under laboratory settings. and Hughes tested it in the field and provided a great more insight into how drift-feeding fishes encounter their world.


For me, Nick Hughes work on Arctic Grayling (Thymallus arcticus) in Alaska was particularly interesting, and his "bigger fish upstream" hypothesis (Hughes and Reynolds 1994, Hughes 1998, 1999) countered my experiences with trout. To have a brief chance to meet and talk to him at a meeting was a highlight.

Bigger fish upstream makes sense for grayling. Unlike trout which are mostly drift-feeders, grayling with their tiny mouths are nearly exclusively drift feeding fishes. As trout get larger, they tend to shift their diet from insects to larger morsels like crayfish and other fishes. Hughes explanation for bigger fish upstream was drift density - that is the amount of food per volume of water - was greater in smaller streams. In particular, Hughes and Dill's (1990) paper got mathematical and started to describe how grayling see and react to prey and how this influences their maximum capture distance. This became the basis for understanding how drift-feeding fishes react to prey.

Much of the research also found a dominance hierarchy where the largest, most dominant fish was in the most profitable position and so on down the line. Smaller fishes were pushed to suboptimal habitats. However, for the smallest of fishes, their swimming ability also means that habitats in slow water are more suitable for them. Many of the early studies looked at relationships between native and non-native fishes - Brown and Brook Trout (Fausch and White 1981) - or hatchery and wild trout (Bachman 1984). Later efforts by Hill and Grossman (1993) and Railsback (i.e. Railsback and Harvey 2002) incorporated bioenergetics models that weighed costs and benefits more explicitly.


As science does, good ideas are kept and modified to better fit new observations. These early models provided a theoretical framework that contiues to serve as the backbone for current investigations. For example, Gowan and Fausch (2002) recognized that the profitability of position was seasonal dependent and Piccolo et al. (2014) recognized that field tests of these theories are generally lacking. Harvey and Railsback (2014) and other researchers recognized that drift-feeding is only part of the story and more active searching is often used by trout and incorporation of active foraging into our models fit the data better. In 2024, Rossi and colleagues used Fausch et al.'s (2002) riverscape concept to describe "foodscapes" of fishes.


For a more detailed - and first-hand - review of the development of these models, read the historical perspective by Kurt Fausch. If you have the time and inclination to read a review of the development of these models, I suggest reading Piccolo et al. 2014. As always, I'm happy to share articles if you want to read them for yourself.

Look for an upcoming post - who knows when... - on what this all means for the angler. But my goal is to make these posts "bite sized" pieces so I'm going to wrap this one up.


Links to External Sources


AI Generated Annotated Bibliographies


Below are AI generated - through ChatGPT's Deep Research and Perplexity, an AI which specializes in scientific research, that were created by prompts I wrote and modified. I include these for those that want to dig deeper into this topic.


Literature Cited / References / Reading List



















Subscribe Form

Thanks for submitting!

©2020 by The Scientific Fly Angler. Proudly created with Wix.com

bottom of page