Choice of Bluebird Nest Box: A Potential for Ecological Traps?

Choice of Bluebird Nest Box: A Potential for Ecological Traps? Julie Perreau and Dr. Kathryn E. Sieving Department of Wildlife Ecology and Conservation, University of Florida, Gainesville FL 32611 The Eastern bluebird has become accustomed to using nest boxes as a substitute for natural cavities. We compared two popular nesting box models in order to determine whether nest box design and temperature influence breeding success and offspring quality, and whether bluebird pairs exhibited any preference for different boxes. Thirty-seven pairs of Gilbertson and Peterson nest boxes were installed throughout the University of Florida campus and monitored weekly over two clutch periods in one breeding season. Temperature buttons were installed in boxes, and offspring data were recorded on the first day of incubation, and 5 and 12 days after hatching. Bluebirds chose both box designs equally, but we detected a significant effect of box type on mean number of hatchlings, mean chick mass (day 5 and 12 post hatching), and number of chicks fledged (first clutches and overall but not for second clutches). The Gilbertson-nesting birds produced more, higher quality young, suggesting that an inappropriate box design could compromise fitness of parents and offspring, therefore box designs should be tested more widely to clarify the best choices for bird conservation.

INTRODUCTION The Eastern bluebird

Purpose and Hypotheses

The Eastern bluebird (Sialia sialis; EABL) is a small-tomedium sized passerine bird in the Thrush family (Turdidae). This species prefers breeding in cavities found on open land or forest edge habitat. In the 1960’s and 1970’s, reduced nest site availability, due to human-driven disturbances such as deforestation and introduced species competition, caused a significant decline in their populations throughout the SE United States. In response, enthusiasts began installing nest boxes, and EABL became accustomed to using them as a substitute for natural cavities (Davis & Roca, 1995).

Few studies have considered the relationship between nest box model preference and breeding behavior or breeding success in birds using them. Therefore, we tested the hypotheses that box type would influence model preference and reproductive performance in Eastern bluebirds. We selected two popular box models in order to inform both bluebird enthusiasts and researchers of any consequences of using these nest box choices for the popular Eastern bluebird. Our predictions included:

Nesting Boxes

 

Nesting boxes provide a unique opportunity to study avian response to environmental variables, while fostering public education about avian ecology and conservation. However, consequences of selection of nest box type for the birds have not been assessed and can be important to nesting success. Factors affected by nest box selection and placement that could influence reproductive performance include box orientation, surrounding vegetation, and variations in standard nest box dimensions (Horn et al, 1996; see also Lockwood, Moulton, & Scott, 1993; see also Rohrbaugh and Yahner, 1997). Also, the climate within nest boxes (temperature and humidity) can influence avian reproduction, and slight variations in microenvironment caused by box design could have significant impacts on growing birds (Navara and Anderson, 2011).



EABL will exhibit preferences for nest box model if presented with a choice of model. Breeding success will vary with different box designs. Temperature is likely to vary between box designs and may be correlated with differences in egg/chick development between the box types.

METHOD Nest Box Design and Installation Thirty-seven pairs of Gilbertson and Peterson nest boxes were installed throughout the University of Florida campus, mainly in agricultural fields and student gardens. Boxes within pairs were installed 30 meters apart from each other, at the same height (on 5 foot poles), facing the same direction, and over bare ground (> 5 feet from vegetation > 3 feet high; Rohrbaugh and Yahner, 1997). Box pairs were > 100 meters apart, to assure independence.

University of Florida | Department of Wildlife Ecology & Conservation 1

JULIE PERREAU AND DR. KATHRYN E. SIEVING

Procedure and Measurements Boxes were monitored for activity at least once a week, from late February until late July, 2013. Upon initial egg sighting, temperature buttons were placed underneath nesting material, and were programmed to record the temperature every 5 minutes, for a total of three weeks during the first clutch, only. Upon clutch completion, the mass (in g) of each egg was recorded. On the 5th day after hatching we recorded hatchling mass, and on the 12th day after hatching we recorded hatchling mass, wing chord, and tarsus length. All abandoned and destroyed clutches were noted. Parametric statistical analyses were conducted in Statistica (Statsoft, Inc., 2010); p-values < 0.05 were taken as significant.

Peterson box throughout the breeding season, despite starting out with similar clutch sizes. This was primarily due to differences in hatching rate that, in turn, led to differences in fledging rate. Box Type

Nests Total

Mean # Eggs

Mean # Hatchlings

Mean # Fledglings

Clutch 1

Gilbertson

13

4.62

4.10*

4.00*

Peterson

11

4.60

2.89*

2.67*

Clutch 2

Gilbertson

5

4.17

3.67

3.67

Peterson

6

4.00

2.50

2.00

* Significant differences (p < 0.05) Table 1: Mean # of Nests, Eggs, Hatchlings, & Fledglings Per Box Type & Clutch Clutch Size 5 4.5

Number of Hatchlings

Number of Fledglings

4.38

4.47 4

3.92

4

Figure 1: Gilbertson box model (left) and Peterson box model (right)

Mean

3.5

2.77

3

2.62

2.5 2

RESULTS

1.5

Nesting Box Selection

0.5

1 0

Both box types were selected equally (Table 1). Out of the 37 box-pairs, bluebirds occupied both box types in 15 boxpairs. In 5 box-pairs only the Gilbertson box was occupied, in 5 box-pairs only the Peterson model was occupied, and 7 box-pairs were unoccupied.

Gilbertson

Peterson Box Type

Figure 1: Mean (± 1 SE) Clutch Size, # of Hatchlings, & # of Fledglings Throughout Season, per Box Type

Patterns in Egg and Chick Mass Eggs, Chicks, and Fledglings Produced We found no difference between the box types in mean number of eggs for either Clutch 1 or 2. Mean number of hatchlings was significantly greater in the Gilbertson boxes for Clutch 1 (F = 8.0, p = 0.01), but not Clutch 2 (F = 1.8, p = 0.27). Mean number of fledglings was significantly greater in the Gilbertson boxes for Clutch 1 (F = 7.2, p = 0.02), but not Clutch 2 (F = 1.8, p = 0.27; Table 1). When using all data pooled, rather than separated by clutch number, mean clutch size did not differ significantly between boxes (F = 2.1, p = 0.24). However, the mean number of hatchlings (F = 7.6, p = 0.01) and mean number of fledglings was significantly different (F = 6.9, p = 0.02; Figure 1). In summary, the Gilbertson box produced significantly more hatchlings and fledglings than the

Mean egg mass was significantly greater in Gilbertson boxes for Clutch 1 (F = 8.0, p = 0.01), but not Clutch 2 (F = 0.9, p = 0.41). The mean 5th day chick mass was significantly greater in the Gilbertson boxes for Clutch 1 (F = 11.8, p < 0.01), but not Clutch 2 (F = 0.3, p = 0.6). The mean 12th day chick mass was significantly greater in the Gilbertson boxes for Clutch 1 (F = 6.5, p = 0.03, Table 2). Using data collected through the entire breeding season, we found that mean egg mass differed significantly between boxes (F = 8.0, p = 0.01). In addition, the mean 5th day chick mass was significantly different (F = 9.3, p < 0.01), and the mean 12th day chick mass was significantly different (F = 6.3, p = 0.02; Figure 2). In summary, the Gilbertson box produced significantly heavier eggs, hatchlings, and fledglings than the Peterson box.


Mean Egg Mass

Mean 5th Day Chick Mass

Mean 12th Day Chick Mass

Clutch 1


Gilbertson

13.23*

82.69*

106.09*

Peterson

12.61*

43.21*

72.06*

Clutch 2

Gilbertson

11.90

54.16

84.38

Peterson

9.8

59.07

59.57

Box Type

* Significant differences (p < 0.05) Table 2: Mean Egg Mass, 5th Day Chick Mass, & 12th Day Chick Mass Per Box & Clutch

Temperature Effects Incubation and hatchling nest temperatures were not statistically different between box types (means, highs, and lows of days, nights, or both days & nights averaged) but measures suggest that Peterson boxes were consistently warmer during Clutch 1. For this analysis, only Clutch 1 data were used (see Procedures and Measurements). Mean Daytime High (C°)

Mean Daytime Low (C°)

Mean Nighttime High (C°)

Mean Nighttime Low (C°)

Gilbertson

31.29

17.43

24.62

15.43

Peterson

32.13

22.98

29.51

21.67

Table 3: Mean Daytime and Nighttime High & Low Temperature (Clutch 1) Clutch Mass

102.14

120

5th Day Brood Mass 100

12th Day Brood Mass

76.58

Breeding Success 68.21

Mean

80 47.53 60 40 20

12.86

11.49

0 Gilbertson

Peterson Box Type

Figure 2: Mean (± 1 SE) Egg Mass, 5th day Chick Mass, & 12th day Chick Mass Throughout Season, per Box Type

Tarsus and Wing Measurements Mean tarsus length did not differ significantly between box types throughout the breeding season (F = 0.6, p = 0.44), but mean wing chord length was significantly larger in Gilbertson boxes (F = 10.4, p < 0.01). 65

61.62

Tarsus Length

Wing Chord

60 55

51.31

Mean

50 45 40 35 30 25

DISCUSSION

20.71

20.54

20

In summary, we found that nest box design has a significant impact on the breeding success of Eastern bluebirds. Although birds nesting in both box types had similar clutch sizes, Gilbertson boxes produced significantly more hatchlings and fledgling chicks in the first clutch and throughout the breeding season overall (Table 1, Figure 1). Furthermore, the mean mass of hatchlings raised in Gilbertson boxes was consistently significantly higher than in Peterson boxes (Table 2, Figure 2). This finding has important implications, because survival of juveniles in their first year, up to breeding age, is known to be positively correlated with fledging mass in many avian species (NaefDaenzer et al, 2001). Our data suggest that box type could, therefore, influence juvenile EABL survival and should be tested before box design recommendations are promoted. Gilbertson-nesting birds produced more, higher quality young, particularly in the first clutch. We expected to see a decline in nesting productivity in Clutch 2 because parental care is energetically expensive for birds and, as a general rule, effort invested declines in later clutches in species that undertake more than one nesting per season (Lack, 1947). In species where both parents invest in young, as is the case for EABL, conflict is higher between progeny in later broods. Within-brood conflict can reduce offspring quality, and this has been linked to the fact that earlier broods receive more parental investment (Parker, 1985) than later, lesser quality broods. This overall relaxation of parental effort could explain the lack of difference between box type productivity in Clutch 2 (Fig. 4).

15 Gilbertson

Peterson

Temperature Influences

Box Type Figure 3: Mean (± 1 SE) Tarsus and Wing Chord Length Grouped by Box Type

Parametric statistical tests (F-tests) comparing daily temperature means, highs, and lows over incubation and



hatchling periods within Clutch 1 detected no significant temperature differences between the box models in Clutch 1. However, with high variance in temperature measures, larger sample sizes may be needed to detect meaningful differences using F-tests. Indeed, non-parametric ranks tests detected consistently higher temperatures in the Peterson boxes (means varied by more than 5 C°). Butler et al. (2009) detected a significant effect of only 0.6 C° on hatching success in kestrel nest boxes, and in Massachussetts, artificially heated nests produced higher quality swallow fledglings than unheated nests (Peréz, et al. 2008). [We note that latitude reverses the sign of temperature influence on nest success; in hot FL, more heat may hurt nestlings, but in colder climates, heat can enhance their growth.] We conclude that observed differences in reproductive success and offspring quality of EABL may be attributed to a temperature effect, though more data are needed to confirm this link. We were careful to place both box types in each pair in shade or in sun to minimize intra-pair variation in temperature. Additionally, both box models have shade overhangs and are well ventilated. Therefore, the consistently higher temperatures that we detected in the Peterson boxes likely derived from the box design. Impact of Box Shape Differences in reproductive success and chick development may be due to space constriction inside nest boxes. Not only was mean chick mass in Gilbertson nest boxes significantly higher at fledging, but hatchlings raised in Gilbertson nest boxes also had significantly larger average wing chords than those raised in Peterson boxes, reflecting more advanced wing development at a stage when mobility is critical to survival. Food restriction during the nestling period has been linked to shorter wing chords at fledging (Lyons and Roby 2011). Whereas general chick growth is already highly advanced after one week, wing growth and feather development is concentrated in the second week of development (Pinkowski, 1974). One explanation is that crowding of older nestlings in narrow-bottomed Peterson boxes could hinder nestling access to food during deliveries by the parents, and may exacerbate intra-brood conflict (competition, fighting) leading to arrested wing growth. Interestingly, nestlings with relatively short wings may restrict their own food intake to lighten the wing-loading at fledging (Wright et al. 2006). Therefore, wing chord (more than body mass) is a reliable indicator of nutritional limitation during development, and we suggest that EABL nestlings may be relatively food-limited in Peterson boxes. Future Work Given that EABL selected both box types equally, this suggests that under certain conditions, a poorly designed nest box could create an “ecological trap”: a resource that is relatively unsuitable or dangerous to a species but that

provides the same cues as one that is adequate or safe to use (Schlaepfer et al, 2002). Our findings suggest that bluebirds were attracted to nest boxes regardless of the design or the reproductive consequences. Therefore, in cases where only inappropriate or poorly designed nest boxes are available for EABL nesting, well-meaning enthusiasts could potentially create a population sink where highly attractive nest boxes reduce reproductive output so severely that the breeding population cannot replace itself. Given that native breeding birds face many human-caused limitations on reproductive success (habitat conversion, feral cats, invasive species, etc.), it is important to insure that resources provided for their benefit do not unknowingly intensify limitations. To prevent this, more rigorous nest box approval procedures may be required. Many box designs are found alongside recommended designs promoted by the North American Bluebird Society (NABS; including Gilbertson and Peterson boxes: www.nabluebirdsociety.org). While we provide evidence that these 2 boxes are equally attractive but cause different success, further work is needed to test all approved box designs side by side. In different regions perhaps only those that are proven to maximize reproductive success should be promoted, especially where a large proportion of the breeding population must rely on nest boxes. REFERENCES Butler, M. W., B. A. Whitman, and A. M. Dufty Jr. (2009). Nest Box Temperature and Hatching Success of American Kestrels Varies with Nest Box Orientation. The Wilson Journal of Ornithology 121(4), 778-782. Davis, W. H., and P. Roca (1995). Bluebirds and Their Survival. Lexington, KY: University of Kentucky. Print. Horn, D. J., M. Benninger-Traux, and D. W. Ulaszewksi (1996). Brief Note: The Influence of Habitat Characteristics on Nestbox Selection by Eastern Bluebirds (Sialia sialis) and Four Competitors. The Ohio Journal of Science 96(3). 57-59. Lack, D. (1947). The Significance of Clutch Size. The Ibis 89(90). 302-352. Lockwood, J. L., M. P. Moulton, and M. A. Scott (1993). Effects of Microhabitat on Nest Box Selection and Annual Productivity of Eastern Bluebirds (Sialia Sialis) in Southeastern Georgia. The Texas Journal of Science, 45(1), 77-85. Lyons, D. E., D. D. Roby (2011) Validating growth and development of a seabird as an indicator of food availability: captive-reared Caspian Tern chicks fed ad libitum and restricted diets. Journal of Field Ornithology 82(1), 88-100. Naef-Daenzer, B., F. Widmer, and M. Nuber (2001). Differential Postfledging Survival of Great and Coal Tits in Relation to Their Condition and Fledging Date. Journal of Animal Ecology 70(5), 730-38. Navara, K. J., and E. M. Anderson (2011). Eastern Bluebirds Choose Nest Boxes Based on Box Orientation. Southeastern Naturalist 10(4): 713-20. Parker, G. A. (1985). Models of Parent-Offspring Conflict v. Effects of the Behaviour of the Two Parents. Animal Behaviour 33. 519-533. Peréz, A. H., D. A. Ardia, E. K. Chad, and E. D. Clotfelter (2008) Experimental heating reveals nest temperature affects nestling condition in tree swallows (Tachycineta bicolor). Biology Letters 11(4). 468-471. Pinkowski, B. C. (1974). A Comparative Study of the Behavioral and Breeding Ecology of the Eastern Bluebird (Sialia Sialis). Thesis, Wayne State University, Dept. of Biology. Rorbaugh Jr., R. W. and R. H. Yahner (1997). Effect of Macrohabitat and Microhabitat on Nest Box Use and Nesting Success of American Kestrels. Wilson Bulletin 109(3). 410-423. Schlaepfer, M. A., M. C. Runge, and P. W. Sherman (2002). Ecological and Evolutionary Traps. Trends in Ecology and Evolution 17(10). 474-480. StatSoft, Inc (2010). \STATISTICA Version 9.1. URL http://www.statsoft.com/. Wright, J., S. Markman, and S. M. Denney (2006) Facultative adjustment of prefledging mass loss by nestling swifts preparing for flight. Proceedings of the Royal Society B 273(1596): 1895–1900.
