Variation in food quality and quantity is the basis of optimal foraging theory

As we discussed earlier in this chapter, animal behavior is a form of phenotypic plasticity and therefore can be considered under the same conceptual framework. Foraging is one of many important behaviors for animals and a great deal of research has been conducted on how animals search for food and must select from a diversity of choices. Because abundance of food items varies over space and time, no single feeding strategy can maximize an animal’s fitness. Hence, feeding decisions represent phenotypically plastic behavior because different feeding strategies represent different behavioral phenotypes. These phenotypes are induced by unique environmental cues and each alternative feeding behavior is well suited to a particular environment but not well suited to other environments. Therefore, the alternative behavioral phenotypes experience fitness trade-offs.

Animals must determine where to forage, how long to feed in a certain patch of habitat, and which types of food to eat. Ecologists evaluate foraging decisions by estimating the costs and benefits of feeding in particular environmental situations. They then compare these estimates to observations of foraging animals to see which strategy provides the highest fitness. Although it would be ideal to measure costs and benefits in terms of survival and reproduction, these components of evolutionary fitness can be difficult to measure. Consequently, ecologists usually look at factors correlated with fitness, such as foraging efficiency. This is based on an assumption that animals able to gather more food in less time should be more successful at survival and reproduction.

104

Animals have four responses to food variation in space and time: central place foraging, risk-sensitive foraging, optimal diet composition, and diet mixing.

Central Place Foraging

Central place foraging Foraging behavior in which acquired food is brought to a central place, such as a nest with young.

When birds feed their offspring in a nest, the chicks are tied to a single location while the parents are free to search for food at a distance. This situation is known as central place foraging because acquired food is brought to a central place, such as a nest with young. As the parent ranges farther from the nest, it finds a greater amount of potential food sources. However, traveling a longer distance increases time, energy costs, and exposure to risk. The animal must choose the amount of time spent gathering food before returning to the nest as well as how much food to bring back with each trip.

Researchers have used these choices to investigate the feeding behavior of European starlings (Sturnus vulgaris). During the summer, starlings typically forage on lawns and pastures for the larvae of craneflies, called leatherjackets. Starlings feed by thrusting their bills into the soft ground and spreading their beak to expose the prey. When they are gathering food for their young, they hold captured leatherjackets in the base of their bill. Researchers predicted that as a starling continued to capture more leatherjackets and to hold them in its bill, it would become more difficult to grab the next one. This is analogous to shopping for items in a grocery store without a cart or a basket; the more items you hold, the harder it is to add another. As a result, the number of prey caught by the starlings should slow down over time. As can be seen in Figure 4.18, the prediction was supported by the data. The shape of the curve shows that the rate of food gathering rises rapidly at first and then, as the starling fills its beak, begins to slow down. We say that the starlings experience diminishing benefits over time.

Figure 4.18 Diminishing benefits over time. The rate of food gathering for the European starling is rapid at first but, as time passes, it experiences diminishing benefits because the amount of prey gathered per unit of time decreases.

The rate at which the parent bird brings food back to its offspring is a function of how much food it obtains and how much time it takes to obtain it. The total time it takes to obtain the food depends on the time needed to fly round-trip to the site that contains the food, known as the traveling time, plus the time spent obtaining the food once the bird arrives at the site, known as the searching time. Figure 4.19 shows a graphical model of how an animal should make decisions as a central place forager. The diminishing benefits line is shown in orange. To this, we can add a fixed traveling time, which is the amount of time the bird needs to get to the feeding area. We can then draw a red line from the origin of the trip to intersect the benefits curve. If we draw the red line at the steepest slope that intersects the orange benefits curve, the two lines cross at the red dot on the figure. This intersection point—which is drawn tangent to the orange benefits curve—represents the highest rate of food capture the bird can obtain, including traveling time. If the starling expressed any alternative behavioral phenotypes—for example, if it stayed at the feeding site for longer or shorter periods indicated by the black dots in the figure—it would have a lower rate of food acquisition.

Figure 4.19 Central place foraging. The optimal rate of foraging for an animal that leaves its nest to find food depends on the time needed to travel to a location that contains food and the time spent feeding once it has arrived. For a given benefits curve (the orange line), the optimal rate of prey capture is found by drawing a straight line from the origin of the trip tangent to the benefits curve. The point of tangency indicates the optimal time that the animal should spend searching and the optimal amount of food it should bring back. Spending more or less time feeding in the location, as indicated by black dots, results in suboptimal amounts of food obtained per unit of time.

Given our understanding of how the starling should forage when the feeding location is at a fixed distance from its nest, how should the bird’s behavior change when the food source is closer or farther away? At sites that are farther away, the bird should spend more time searching for food and bring back more food to help offset the extra travel time. In contrast, as travel time decreases for sites that are closer to the nest, the bird should spend less time searching the site for food and bring back less food. Recall the example of the grocery store. If the store happened to be across the street from your house, you would probably make frequent trips, spend a short amount of time searching for food, and bring back a few items on each trip. If the store were an hour’s drive away, you would likely make fewer trips, spend a longer time searching for food, and bring back an armload of items on each trip. Like the starling, these decisions improve your efficiency in bringing back food.

105

To what extent do organisms actually forage optimally? Researchers addressed this question in a clever experiment. They trained starlings to visit feeding tables that offered mealworms through a plastic tube at precisely timed intervals. A starling would arrive at the table, pick up the first mealworm, and then wait for the next mealworm to be delivered. Each successive mealworm was presented at progressively longer intervals, mimicking the longer intervals at which a starling would catch leatherjackets as its beak became increasingly full. The researchers set up feeding tables at different distances from the starling nests and observed how many mealworms a starling picked up before it departed back to its nest. The predicted number of prey brought back to the bird’s nest, shown as a blue line in Figure 4.20, agrees with the actual number observed in the experiment, represented by the red data points.

Figure 4.20 Predicted versus observed prey capture for a central-place forager. Based on an optimal foraging model, researchers predicted that longer travel times would cause starlings to return to their nests with a larger number of mealworms. The researchers offered mealworms to starlings on tables that were located at different distances from their nests. The observed number of mealworms brought back to the nests show agreement with the predictions.
After A. Kacelnik, Central place foraging in starlings (Sturnus vulgaris), I. Patch residence time. Journal of Animal Ecology 53 (1984): 283–299.

Risk-Sensitive Foraging

Risk-sensitive foraging Foraging behavior that is influenced by the presence of predators.

Our predictions of how animals should forage have assumed that animals are simply maximizing their rate of energy gain. However, most animals have other considerations, including predators. Because the act of feeding puts most animals at risk for predation, they must consider this danger when making their foraging decisions. Animals that incorporate the risk of predation into their foraging decisions are said to practice risk-sensitive foraging.

The creek chub is a fish that faces the common challenge of finding its lunch rather than becoming lunch for a predator. Small creek chubs feed on tubifex worms and they prefer to feed in locations that have more worms. But what if locations containing more worms also contain more predators including cannibalistic larger creek chubs? How much food would it take to entice the small creek chubs to feed in the riskier location? To address this question, researchers placed small creek chubs into an artificial stream that contained a refuge from predators in the middle section of the stream. On one end of the stream, the researchers placed a large creek chub and a low density of worms. On the other end of the stream, they placed two large creek chubs and manipulated different densities of worms. As you can see in Figure 4.21, when the opposite end of the stream had two large creek chubs, the small creek chubs would not move to the side containing two predators until that side contained three times as much food.

Figure 4.21 Risk-sensitive foraging. The sensitivity of young creek chubs (Semotilus atromaculatus) to food density and predators was tested in artificial streams. All streams had one side that contained one predatory adult creek chub and a low density of food (0.17 worms/cm2). (a) When the right end of the stream contained two predators and the same low density of food, the young chubs moved to the left side. (b) When the right end contained two predators and a medium food density (0.17 worms/cm2), the young chubs still moved to the left side. (c) Only when the side with two predators contained either a high food density (0.50 worms/cm2) or (d) very high food density (1.0 worms/cm2) did the young chubs move to the right side of the stream.
After J. F. Gilliam and D. F. Fraser, Habitat selection under predation hazard: Test of a model with foraging minnows, Ecology 68 (1987): 1856–1862.

106

107

Optimal Diet Composition

Most animals do not consume a single food item but choose from a range of food items. For example, consider the food choices of coyotes living in the western United States. In Idaho, the coyote can consume a variety of prey species, including small prey such as voles (Microtus montanus), medium prey such as cottontail rabbits (Sylvilagus nuttallii), and large prey such as jackrabbits (Lepus californicus). Larger prey species provide larger energy benefits to the coyote, but require more time and energy to subdue and consume. Given these options, the coyote has to decide which prey species it should pursue and which prey species it should ignore.

Handling time The amount of time that a predator takes to consume a captured prey.

To determine the optimal food decision we must balance the energy obtained from the prey and the handling time—the time required to subdue and consume the prey. Assuming the same amount of handling time for each choice, the optimal decision for the predator will depend on the energy obtained from each prey and the abundance of the prey. The optimal decision can change if handling time is not equal. In this situation, we need to consider the amount of energy gained per unit time for each prey species. We can do this by dividing the energy benefit of a prey item by its handling time. When we do this, we sometimes find that the smallest prey should be consumed because, although they provide a smaller energy benefit than larger prey, their low handling time can provide the predator with the highest energy gain per unit time. In the case of the coyotes, researchers have found that although jackrabbits require more effort to catch and consume than cottontail rabbits or voles, jackrabbits also provide a larger energy benefit, so the coyotes should rank jackrabbits as the most profitable food item, followed by cottontail rabbits and then voles.

Once we know how different food items compare in terms of energy gained per unit of handling time, we can make a number of predictions. For example, the predator should always eat the prey species that provides the highest energy benefit; if this prey is abundant, it is the only prey that the predator should consume. This strategy maximizes the animal’s energy gain. However, if this highest energy prey is rare and the predator’s energy needs are not met, the animal should include less profitable items in its diet. Prey species of very low energy value should never be included in the diet unless all higher energy prey are scarce. In the case of the coyote, researchers have found that the coyotes appear to be making optimal diet choices. The coyotes always consumed the jackrabbits regardless of their abundance. However, when jackrabbits became less abundant, the coyotes would increase their consumption of cottontail rabbits and voles.

Diet Mixing

Some foragers consume a varied diet because one type of food might not provide all of the necessary nutrients. Humans, for example, can synthesize many amino acids, but other amino acids—known as essential amino acids—can only be obtained from one’s diet. A diet of only rice or only beans does not possess the complete set of essential amino acids needed by humans. However, a diet that combines rice and beans contains all of the required essential amino acids because each contains the essential amino acids that are missing in the other.

The benefits of diet mixing have been demonstrated using nymphs (immature stages) of the American grasshopper (Schistocerca americana). As you can see in Figure 4.22a, grasshopper nymphs grew faster when fed a mixture of kale, cotton, and basil than when they were offered any one of these food plants alone. The effect was even more pronounced on lower-quality, natural food plants, such as mesquite and mulberry: nymphs with mixed diets grew almost twice as fast as those feeding on either one of these plant species alone, shown in Figure 4.22b. Similar results were obtained on artificial diets that were low in either protein or carbohydrates, both of which are required for proper growth: grasshoppers with mixed diets grew more rapidly. Based on these data, we might predict that when given a choice, these grasshoppers would decide to forage on a mixed diet to improve their fitness.

Figure 4.22 Mixed diets. Young grasshoppers grow faster on mixed diets than on any single diet, regardless of whether comparisons were made using (a) crop plants or (b) natural plants. In general, mixed diets supply a more complete range of nutrients needed by animals than single diets. Error bars are standard errors.
After E. A. Bernays et al., Dietary mixing in a generalist herbivore: Tests of two hypotheses, Ecology 75 (1994): 1997–2006.

108

Throughout this chapter, we have seen that organisms commonly experience spatial and temporal variation in their environment. In response to this variation, many have evolved the ability to produce multiple phenotypes from a single genotype. The strategy of using multiple phenotypes—including changes in morphology, physiology, or behavior—is effective when there are trade-offs such that no single phenotype performs well in all environments. The evolution of phenotypic plasticity is common among all groups of organisms on Earth, wherever there are reliable environmental cues.

ECOLOGY TODAY CONNECTING THE CONCEPTS

RESPONDING TO NOVEL ENVIRONMENTAL VARIATION

Elevated CO2 experiment. Tall towers at the Duke University Forest have been emitting CO2 into the atmosphere for several years and researchers have been tracking the effects on the plants.
Photo by Jeffery S. Pippen, http:people.duke.edu/∼jspippen/nature.htm.

Ecologists have a good understanding of phenotypically plastic adaptations to environmental variation that has been present for hundreds of thousands of generations, long enough to evolve an appropriate phenotypic response mechanism. But how do organisms respond to more recent environmental variation?

One of the most profound changes in our environment has been the global increase in atmospheric CO2. In 1958, Charles Keeling began recording atmospheric CO2 concentrations atop 3,400-m-high Mauna Loa on the island of Hawaii. Keeling wanted to determine whether anthropogenic emissions were increasing the concentration of CO2 in the atmosphere. At the time he began his study, scientists had no accurate long-term measurements of atmospheric CO2 concentrations. In 1958, the CO2 concentration was about 316 parts per million (ppm; 316 CO2 molecules per million molecules of air, mostly nitrogen and oxygen). During the subsequent decades, the concentration of CO2 in the atmosphere has increased dramatically, rising to 352 ppm by 1990 and 395 ppm by 2012, with no sign of leveling off. Other research indicates that this concentration of CO2 has not been present on Earth for at least the past 10,000 years. As demand for energy and agricultural land increases, the concentration of CO2 is expected to reach 500 to 1,000 ppm by the year 2100.

109

How will organisms respond to such a change in their environment given that they have not experienced such high concentrations of CO2 in the past 10,000 years? To address this question, scientists have been conducting large outdoor experiments in which tall towers emit CO2 gas over forests and croplands to cause an elevation in CO2 ranging from 475 to 600 ppm. Averaged across several such experiments that were initiated in the mid-1990s, researchers have found that elevated CO2 causes an increase in the rate of photosynthesis by 40 percent. In addition, because plants open their stomata to obtain CO2, the elevated CO2 concentrations allow plants to keep their stomata closed more often, resulting in a 22 percent reduction in transpiration of water. All of this translates to improved plant growth. Plants experiencing elevated CO2 experienced a 17 percent increase in the growth of their shoots and a 30 percent increase in the growth of their roots.

Changes in atmospheric CO2 over time. Measurements on the island of Hawaii have shown that CO2 concentrations have continuously increased during the past 50 years.
After http://www.esrl.noaa.gov/gmd/ccgg/trends/

These growth responses represent averages across a variety of species, but not all species responded in the same manner. For example, growth improved in C3 plants but not in C4 plants. Researchers hypothesize that because the C4 pathway of photosynthesis already pumps high concentrations of CO2 into the bundle sheath cells of the leaf, higher atmospheric concentration of CO2 has little additional effect. Given that most plant species use the C3 pathway, most plants will experience higher growth unless other nutrients become limiting or if herbivory of the plants also increases and causes an increased loss of plant tissues. On the other hand, C4 plants, which include corn, sugar cane, and many other important crops, are not expected to grow any faster as humans continue to elevate the concentration of CO2 in the atmosphere.

The change in CO2 concentrations is just one example of the many anthropogenic changes occurring on Earth today. Most organisms have flexible phenotypes that have been shaped by natural selection in response to past environmental variation. Organisms facing novel environmental variation from anthropogenic causes may be able to make use of these existing adaptations and they may also experience continued evolution for new types of flexible phenotypes. However, many other types of anthropogenic impacts—such as air and water pollution—may be far outside the range of historic environmental variation for a population. As a result, populations may not possess phenotypically plastic strategies that will allow them to perform well when facing these types of anthropogenic impacts.

SOURCES: Jaub, D. 2010. Effects of rising atmospheric concentrations of carbon dioxide on plants. Nature Education Knowledge 1: 21.

Ainsworth, E. A., and S. P. Long. 2005. What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytologist 165: 351–372.

110