H O W- T O - D O - I T

Using Artificial Nests To Study

Nest Predation IN BIRDS

J A M E S R. B E LT H O F F

N

est predation occurs when either eggs or nestlings within nests of birds are destroyed or consumed by predators. Because of the loss of their offspring, such behavior reduces the reproductive success of the parent birds, and their lifetime fitness declines. Untold numbers of bird nests are depredated each year, and ecologists and wildlife managers are interested in the factors that affect the likelihood of nest predation or success.

James R. Belthoff is Professor of Biology at Boise State University, Boise, Idaho 83725; e-mail: jbeltho@boise state.edu.

Why Study Predation on Artificial Bird Nests? Studies with artificial bird nests are useful for illustrating basic ecological principles, such as how predation occurs differently in different habitats (Rudnicky & Hunter, 1993; Leimgruber, McShea & Rappole, 1994), effects of land use changes such as those resulting from habitat fragmentation (Bayne & Hobson, 1997) on predation, and the effects of distance from edge and edge type on predation (i.e., the so-called “edge effects,” where predation increases near ecotones [Paton, 1994]). The coevolution between predators and their prey also is of interest NEST PREDATION IN BIRDS

105

to biologists and can be examined with artificial nests. This coevolution sets up an “arms race” between the predators and their prey, as natural selection increases the ability with which predators detect and capture prey, but selection also improves the prey’s capacity to avoid detection (Krebs & Davies, 1993). Because many organisms prey on bird nests, including reptiles (e.g., snakes), mammals (ranging in size from small mice to larger carnivores), and many types of birds, there is strong selection on nesting birds to place nests in areas that will elude detection (such as areas with greater cover, in cavities in trees), to distribute nests in spaced rather than clumped distributions, or to cryptically color their eggs. Each of these factors can be examined in studies using artificial bird nests (see Major & Kendal, 1996).

Figure 1. In this article, I describe simple and effective field Artificial nest containing three eggs made of clay designed to resemble that of a Brewer’s Sparrow exercises that demonstrate during a study of nest predation. factors affecting predation on bird nests. With instructor guidance, students in stores or their wholesale suppliers, such as high school biology or college-level biology, ecology, Howell’s Home and Floral (www.howellsonanimal behavior, wildlife management or ornithology line.com; Portland, OR) and resemble those of laboratory courses can collaborate to design field expermany small passerine birds (Figure 1). iments related to nest depredation. Hypotheses about • Plasticine modeling clay for molding eggs (Van nest location (ground, shrub, tree branch, cavity) or Aken International, Rancho Cucamonga, CA; amount of cover, egg number or color, habitat type, or multiple colors are available from local craft natural vs. disturbed environments can be formulated stores) in ~0.45 kg (1 lb) packages. For small and examined. I have used these exercises in introducpasserine eggs (e.g., 1-2 cm in length), each packtory animal behavior and conservation laboratories but, age makes approximately 200-400 eggs. with little or no modification, they are suitable for an array of other courses within the biological sciences. • 26-gauge wire for securing nests to substrate

Materials The following materials will be needed for this exercise: • Artificial bird nests to resemble those of avian species in your area. Nests 7-10 cm (3-4 in) in diameter are commonly available from craft 106

THE AMERICAN BIOLOGY TEACHER, VOLUME 67, NO. 2, FEBRUARY 2005

• 50 m or 100 m measuring tapes for laying out transects or measuring heights or other aspects of artificial nests • Small plastic vials for transporting depredated eggs to laboratory for follow-up analyses of beak and tooth impressions left by predators (optional)

• Dissecting (10X) microscope for examining tooth and beak marks on clay eggs (optional) • Pin flags or plastic flagging tape to mark the general location of artificial nests deployed (optional). In general, these markers should be placed some distance from the nests themselves to avoid cuing predators to the nests. • Vinyl gloves for making and deploying nests to eliminate human scent contamination

Experimental Design Egg Construction Eggs of appropriate size and shape are molded from the modeling clay by hand (Figure 1). Vinyl gloves are used to help prevent human scent from being transferred to eggs, as human scent could alter rates of predation. A guide, such as Baicich & Harrison (1997), is helpful for providing descriptions of eggs for common species in your area.

Nest Deployment Artificial nests can be deployed along linear transects, such as every 10 m along a 100 m transect, or they can be distributed within replicated study plots of

desired size. Generally, it is best to secure the artificial nests to the particular substrate (shrub, tree branch, ground) so wind or the predators themselves do not dislodge nests. We use 26-gauge wire, which takes only a moment to weave through the nest material and around the branch of a shrub or tree.

Some Factors & Questions of Interest Effects of Different Colored Eggs Do white eggs experience more predation than eggs of different colors, including colors that are more cryptic?

Effects of Egg Number Do nests with only one or a few eggs attract fewer predators and end up successful more than nests with more eggs?

Effects of Nest Height Do nests closer to or on the ground experience greater predation than those higher in the shrub or tree canopy?

NEST PREDATION IN BIRDS

107

The color of clay available for making eggs

What To Let Them Determine (With Your Supervision) What are the hypotheses (null and alternate) being examined?

Bird beak marks Mammal teeth marks

Figure 2. Two clay eggs that have been disturbed by predators.The egg on the left has teeth marks characteristic of a mammalian nest predator, while the egg on the right has impressions left by the beak of a bird. Both types of marks are characteristic of nests that students would categorize as depredated. Effects of Nest Distribution Do nests in clumped distributions experience greater predation than those that are randomly spaced?

Table 1. Results of artificial nest study to test the null hypothesis that nest success and predation are independent of nest position (on the ground or in the canopy of a sagebrush shrub).Values represent the number of nests in each category.

How many nests to deploy and in what configuration How many eggs to place in each nest The criteria used to determine whether a nest is successful or failed

General Procedure

Our laboratory periods are three hours in duration, and this exercise generally can be completed in two periods. During the first period, N E S T F AT E students discuss and formulate specific hypotheses to be Nest Position Successful Failed examined and then develop Effects of Edge Ground 15 28 the experimental design to be Do nests closer to used, which takes 30-40 minShrub 8 36 the edge of a habitat utes. We then visit the field patch experience site, which in our case is greater predation, about 20 minutes from camperhaps because of edge effects? pus, to deploy nests. The travel time is also used for modeling the clay eggs, although I usually solicit a few Effects of Predator Type student volunteers to arrive at class approximately an hour early to help me get this started. Students work Are birds, mammals, or other groups more likely in groups of three to deploy the nests: One student to prey on ground nests or nests within other records the data (nest number and location, nest treatment groups? This can be assessed through height, color or number of eggs, and other factors spefollow-up examination of marks in clay eggs from cific to the hypothesis being examined), another flags depredated nests. the nest locations so that nests can be relocated, and the third secures the nests and places the eggs. Spring What To Tell the Students is the best time for this exercise, as birds are building and tending nests, and presumably their predators How many nests are available for their group are accustomed to finding nests at this time, but we have conducted the exercise in the fall with success as The basic approach to deploying nests, such as well. After one week, we inspect the nests for signs how to place the eggs within them and how to of predation. secure nests to substrate

108

THE AMERICAN BIOLOGY TEACHER, VOLUME 67, NO. 2, FEBRUARY 2005

Collection of Data Students relocate and retrieve each of the nests and score them as either successful or failed. Although students may discuss and establish various criteria, we consider intact nests, where all eggs are present and show no signs of disturbance by potential predators, to be successful. Failed nests are those in which eggs exhibit teeth or beak marks or in which eggs have been either partially or completely removed. Tooth and beak marks are sometimes helpful for determining the type of predator that visited the nest (Figure 2). We use small, plastic vials to store marked eggs for examination under a dissecting microscope in the laboratory, which sometimes helps further decipher predator type. Some researchers deploy automated cameras at nests to more accurately determine the identity of predators, but that is likely beyond the scope of a laboratory exercise for introductory classes.

Results Nest Position We recently obtained the following results concerning depredation of artificial nests in relation to nest position in a sagebrush (Artemisia tridentata) steppe habitat in which artificial nests were placed on the ground (i.e., at the base of a shrub or near other suitable cover) or in the canopy of the shrub (at approximately the height of each student’s thigh). Nests were 7.6 cm in diameter and contained 3 eggs modeled from light green clay (Figure 1) to resemble those of Brewer’s Sparrows (Spizella breweri), which is a common sagebrush nesting species in our area. Students deployed 87 nests along 100 m transects at 10 m intervals, of which 64 (73.6%) were unsuccessful because of predation, and 23 (26.4%) were judged as successful (Table 1). Evidence of bird (beak marks) and small mammal (teeth and claw marks) predation was detected from impressions in eggs similar to those in Figure 2.

Data Analysis We used Chi-Square for 2 x 2 contingency table to ask whether fate (successful or failed) is independent of the particular treatment factor. More advanced classes may use logistic regression, generalized linear models, or Cox Regression (survival analysis) analysis to examine nest fate in relation to multiple factors of interest or any potential interactions among them. For the data in Table 1, there was no difference in predation between shrub nests and ground nests (contingency analysis: X2 = 3.12, P = 0.077).

NEST PREDATION IN BIRDS

109

Clumped vs. Spaced Distribution A second study examined the proportion of nests depredated when clumped vs. scattered within 100 m x 100 m plots within sagebrush steppe habitat. Clumped nests were placed within 5 m of a neighbor whereas scattered nests were greater than 20 m from the nearest neighbor. Each plot (n = 16) contained 10 ground nests. We calculated the percentage of nests in each plot that were depredated, and then compared average rate for plots in both treatments using a Wilcoxon ranked sums test (Zar, 1999). In this case, the null hypothesis of no difference in proportion of nests depredated was rejected, as the mean proportion of failed nests was greater for clumped nests (0.82 ± 0.16 [SD]) than in nests spaced further apart (0.51 ± 0.08 SD; Wilcoxon test, z = 3.20, P = 0.001).

Discussion of Experimental Results & Educational Value Results of artificial nest experiments often show that predation rate varies as a function of nest placement, habitat type, land use, and other factors. Such results act as a springboard for discussion of how birds may better conceal their nests, how the search paths of predators may differentially encounter clumped vs. scattered nests, or how egg color might increase or decrease the visibility of nests to predators. We also discuss student opinions about how realistically the artificial nests estimate actual predation rates. These discussions often revolve around the fact that actual nests would have live, adult birds to (1) defend the nests against predators which may reduce predation, or (2) cue predators to the whereabouts of nests, which may simultaneously increase rates. We also discuss how well the students “hid” the nests (i.e., how much cover there was) and if they believe this was a factor in their chances of being depredated. The exercises are suitable for all levels of classes because they offer the flexibility to make them as complex as necessary. Beginning classes in biology, such as in high schools, can examine one factor’s effect on predation rate, such as nest position or nest distribution, as illustrated above. More advanced classes, such as upper level college biology classes, can discuss multiple hypotheses, issues related to experimental design, and more complex data analyses. In either case, my experience has been that students are excited to construct the eggs from clay, deploy the nests, and eagerly examine their fate upon return trips to the study area.

actual eggs increase the realism of the experiment, recent research indicates that quail eggs may underestimate predation by small mammals such as deer mice (Peromyscus maniculatus), because the shells of quail eggs are larger and thicker than eggs of many passerine birds (see Roper, 1992; Bayne & Hobson, 1999). Nonetheless, quail eggs offer a viable alternative for the experiments described, and they can be obtained from a variety of poultry farms (e.g., http://www.north west-gamebirds.com/eggs.htm). Alternatively, brown chicken eggs can be used to resemble larger gallinaceous birds that nest on the ground, such as grouse or pheasants.

Acknowledgments I thank Chris Canon, Catherine Rideout, and Rex Sallabanks for assistance in obtaining materials for these studies and Ian Robertson for suggesting this article. Two anonymous reviewers also provided comments that improved the manuscript.

References Baicich, P.J. & Harrison, C.J.O. (1997). A Guide to Nests, Eggs, and Nestlings of North American Birds, 2nd Edition. New York: Academic Press. Bayne, E.M. & Hobson, K.A. (1997). Comparing the effects of landscape fragmentation by forestry and agriculture on predation of artificial nests. Conservation Biology, 11, 1418–1429 Bayne, E.M. & Hobson, K.A. (1999). Do clay eggs attract predators to artificial nests? Journal of Field Ornithology, 70, 1–7. Krebs, J.R. & Davies, N.B. 1993. An Introduction to Behavioural Ecology, 3rd Edition. Oxford, United Kingdom: Blackwell Scientific. Leimgruber, P., McShea, W.J. & Rappole, J.H. (1994). Predation on artificial nests in large forest blocks. Journal of Wildlife Management, 58, 254–260. Major, R.E. & Kendal, C.E. (1996). The contribution of artificial nest experiments to understanding avian reproductive success: a review of methods and conclusions. Ibis, 138, 298–307. Paton, P.W.C. (1994). The effect of edge on avian nest success: how strong is the evidence? Conservation Biology, 8, 17–26. Roper, J.J. (1992). Nest predation experiments with quail eggs: too much to swallow? Oikos, 65, 528–530.

Additional Follow-up Exercises & Questions

Rudnicky, T.C. & Hunter, M.L. (1993). Avian nest predation in clearcuts, forests, and edges in a forest-dominated landscape. Journal of Wildlife Management, 57, 358–364.

Students can discuss the pros and cons of using quail eggs rather than clay eggs. For example, while

Zar, J.H. (1999). Biostatistical Analysis, 4th Edition. Saddle River, New Jersey: Prentice Hall.

110

THE AMERICAN BIOLOGY TEACHER, VOLUME 67, NO. 2, FEBRUARY 2005