EFFECTS OF FOOD SUPPLEMENTATION AND HABITAT SELECTION ON TIMING OF LESSER KESTREL BREEDING

Transcription

1 Notes Ecology, 83(3), 2002, pp by the Ecological Society of America EFFECTS OF FOOD SUPPLEMENTATION AND HABITAT SELECTION ON TIMING OF LESSER KESTREL BREEDING JOSÉ MIGUEL APARICIO 1 AND RAÚL BONAL 2 Museo Nacional de Ciencias Naturales (CSIC), Departamento de Ecología Evolutiva, J. Gutiérrez Abascal 2, E Madrid, Spain Abstract. Numerous experimental studies providing extra food have concluded that food availability at the beginning of the breeding season constrains the start of egg-laying for female birds (Food Supply Hypothesis) because supplemented females usually lay earlier than nonsupplemented ones. This conclusion has recently been questioned because food addition studies may be confounded by ordered habitat selection. Ordered habitat selection occurs when territories or nests provided with extra food are chosen by individuals of higher quality that may be able to initiate breeding early, regardless of food supply (Habitat Selection Hypothesis). To test these two hypotheses, we performed an experiment using the Lesser Kestrel (Falco naumanni). To reveal effects of ordered habitat selection, extra food was provided to half of the nests in two colonies in which the other nests remained unsupplemented (mixed colonies). We provided extra food to all nests in two colonies (allfed colonies) and to no nests in three colonies (all-unfed colonies). In these colonies, ordered habitat selection could not occur because all nests received equal treatment. In mixed colonies, fed pairs laid earlier than unfed ones. In contrast to the prediction of the Habitat Selection Hypothesis, there was no significant difference in mean laying date between unfed pairs of mixed colonies and pairs in all-unfed colonies, or between fed pairs of mixed colonies and pairs in all-fed colonies. Moreover, laying date was significantly earlier in all-fed than in all-unfed colonies. Therefore, the results support the Food Supply Hypothesis and refute the Habitat Selection Hypothesis. Key words: egg laying; Falco naumanni; food limitation; food supply; habitat selection; laying date; Lesser Kestrel; Spain; supplementary food; timing of breeding. INTRODUCTION The timing of reproduction has been related to several components of fitness in a wide variety of organisms. Early breeders normally produce more offspring of better condition (e.g., Cavers and Steel 1984, Koenig and Albano 1987, Daan et al. 1989, Dobson and Myers 1989, Landa 1992, Schultz 1993, Olsson and Shine 1997) that, typically, have a higher chance of surviving and reproducing than those produced by late breeders (Perrins 1970, Harris et al. 1992, Ydenberg et al. 1995, Brinkhof et al. 1997). Hence, one may ask why late breeders do not reproduce earlier. For birds, Perrins (1970) argued that females should start to lay as soon Manuscript received 27 June 2000; revised 2 April 2001; accepted 4 April Present address: Instituto de Investigación en Recursos Cinegéticos (CSIC, UCLM, JCCM), Ronda de Toledo s/n, E Ciudad Real, Spain. jmaparic@irec.uclm.es 2 Present address: Departamento de Ciencias Ambientales, Facultad de Ciencias del Medio Ambiente, Universidad de Castilla-La Mancha, E Toledo, Spain. as they are physiologically capable, and that interindividual differences in the timing of breeding could be caused by differential acquisition of food. In contrast, some authors maintain that interindividual differences may be adaptive (Daan et al. 1989, Schultz et al. 1991, Aparicio 1994, 1998), and food availability may be a cue to trigger the start of egg-laying rather than a constraint on the production of eggs (Aparicio 1998). Both of these hypotheses (hereafter, the Food Supply Hypothesis), however, consider food supply to be the main factor affecting the onset of reproduction. In support of the Food Supply Hypothesis, numerous experimental studies have shown that females supplied with extra food laid eggs earlier than controls (reviews in Martin 1987, Arcese and Smith 1988, Boutin 1990, Svensson 1994). The conclusions from these experiments have recently been questioned by Kelly and Van Horne (1997), who argued that territories provided with extra food might be settled by individuals of higher quality that might be able to initiate breeding earlier, independently of the experimental treatment. There- 873

2 874 NOTES Ecology, Vol. 83, No. 3 fore, ordered habitat selection (hereafter, the Habitat Selection Hypothesis) might confound the effects of supplemental food. To distinguish between these two hypotheses, Kelly and Van Horne (1997) proposed that supplemental food should be provided to some territories interspersed with other territories not supplemented with food, to allow ordered habitat selection correlated with experimental treatment. Laying dates from these territories should be compared with those of single-treatment reference populations. In this study, we conducted an experiment similar to that suggested by Kelly and Van Horne (1997) using the Lesser Kestrel (Falco naumanni). This falconiform species is a colonial breeder. We placed food in some kestrel nests, but not others, within two different colonies ( mixed colonies ). In these mixed colonies, the Habitat Selection Hypothesis could operate through nest selection, because high-quality individuals could choose the nests supplied with extra food. At the same time, we used other colonies as reference populations in which all pairs received the same treatment (either all-fed colonies or all-unfed colonies ), thus precluding the effects of ordered nest selection. Several predictions can be tested using this experimental design (Fig. 1). Two are particularly important because they separate the effects of food supply from those of ordered nest selection on the timing of breeding. (1) If supplemental food produces an advance of laying date through ordered nest selection, one would expect unfed pairs in mixed colonies to lay later, on average, than pairs breeding in all-unfed colonies. (2) If food supply has a direct effect on laying date, one would expect females in all-fed colonies to lay earlier, on average, than those in all-unfed colonies. METHODS Study area and species This study was conducted on a Lesser Kestrel population located in La Mancha, provinces of Ciudad Real and Toledo, central Spain (39 20 N, 3 15 W). The study area was 1000 km 2 and was set in a plain used for agriculture, cultivated mainly with barley, wheat, and vineyards. Lesser Kestrels form breeding colonies in abandoned field houses and nests are usually under tiled roofs or inside holes in the walls. Lesser Kestrels are migrants in La Mancha; they overwinter in Africa and most of them arrive at the breeding grounds in March. Both males and females are highly philopatric: 60% of surviving young and 90% of surviving adults return to the same colony in which they were born or previously bred (Negro et al. 1997; J. M. Aparicio unpublished data). Movements between colonies normally occur only after a complete reproductive failure in the previous year (Aparicio FIG. 1. Predictions of the Habitat Selection and Food Supply hypotheses for laying date at Lesser Kestrel colonies in Spain under different feeding treatments: all-fed (all nests provided with extra food), all-unfed (no food provided), and mixed colonies (extra food placed in only half of the nests). Dark and light lines represent pairs with and without extra food, respectively. In mixed colonies, dotted lines depict the sum of fed and unfed pairs. Vertical lines represent mean laying dates. The Habitat Selection Hypothesis predicts that, in mixed colonies, high-quality individuals will occupy nests provided with extra food; unsupplemented nests will be occupied by low-quality kestrels. Thus, in these colonies, fed pairs should lay earlier than unfed pairs due to their higher quality, independently of extra food. Because fed pairs of mixed colonies are of higher quality, they should lay earlier, on average, than pairs in all-fed colonies, whereas unfed, lowquality mixed-colony pairs should lay later than all-unfed colony pairs. Pooling all pairs, there should be no differences in laying date between all-fed, all-unfed, and mixed colonies. The Food Supply Hypothesis holds that food availability before laying is the main factor in timing of breeding. Thus, pairs in all-fed colonies should lay significantly earlier than those in all-unfed colonies. Within mixed colonies, fed pairs should bred earlier than unfed pairs, advancing their laying dates to the same degree as all-fed pairs. There should be no differences between unfed pairs of mixed and all-unfed colonies. If both hypotheses are true, the advance of laying date should be greater in mixed than in all-fed colonies in relation to their unfed controls (unfed pairs of mixed and all-unfed colonies, respectively).

3 March 2002 NOTES ). Within colonies, competition for nests sites appears to be strong. The earliest kestrels arrive at the colonies two months before reproduction, when feeding conditions are still quite poor. At the beginning of the season, kestrels normally remain on or near the nest 65% of the daylight hours, or even more if there is an intruder attempting to occupy it. Disputes between pairs are frequent during settlement, but not during other periods. These facts suggest that nest choice normally take place within each colony rather than between colonies. Lesser Kestrels feed mainly on insects, but also prey on small vertebrates such as lizards, birds, and rodents, especially during courtship and reproduction (Franco and Andrada 1974). An increasing proportion of a female s feeding requirement is supplied by her mate beginning from 15 d before egg-laying and ending once the clutch is completed (Donázar et al. 1992). Kestrels lay one clutch per year between mid-april and the end of May, and clutches are normally composed of 4 5 eggs (range 3 6 eggs). Feeding experiment In 1999, we performed an experiment with supplementary food, using seven Lesser Kestrel colonies. We chose colonies that had 10 breeding pairs in previous years to ensure that we would have an adequate sample size within each colony. Colonies were randomly assigned to one of the following three treatments. Two colonies received a mixed treatment (mixed colonies) in which only half of the nest sites were supplemented, the other half serving as unfed controls. In two colonies, all nests were supplied with extra food (all-fed colonies), and in the remaining three no food was provided (all-unfed colonies). We were able to locate potential nest sites before the experiment began because nest sites are reused every year, although usually by different pairs, as only 2 3% of kestrels breed in the same nest as in the previous year (J. M. Aparicio, unpublished data). Within the mixed-colonies treatment, supplemented nests were randomly chosen prior to bird settlement. Food supplementation began in the first week of March when the first kestrel was seen in the study area. Extra food consisted of day-old cockerel chicks (35 40 g) that we placed within each experimental nest to decrease the chances that control pairs would gain access to food. We checked the food every two days, and when it was regularly consumed, we alternately provided one and two chicks every two days. Kestrels normally regurgitate pellets early in the morning from within or beside their nest. We verified that fed pairs were eating extra food, and control kestrels were not: yellow pellets (indicative of the presence of cockerel chick s feathers) were always found at experimental nests but never at control nests. Each potential nest was monitored at least weekly, beginning in the last week of February, to determine its occupation; from mid-april, each nest was monitored every two days to record laying date and clutch size. Kestrels normally lay an egg every two days. Hence, when the first egg was found, we revisited the nest the following day to determine the date of laying more accurately. Kestrels were generally trapped by hand during incubation, measured, and individually marked with metallic and colored plastic rings. In 1999, we caught 50% of the breeding kestrels, and 9.2% (12/130) of these birds had been banded in previous years. To estimate their age, we assumed that birds captured for the first time were in their first year if they had yearling plumage, or in their second year if they had adult plumage. All statistical analyses were performed using SPSS and parametric tests. We verified that data were normally distributed using Kolmogorov-Smirnov tests (in all cases, P 0.46). The means of variables such as age of kestrels, date of nest occupation, and laying dates were compared between colonies and colony treatments using one-way ANOVAs. All significance levels are for two-tailed tests. RESULTS Breeding performance in natural conditions Four of the seven colonies included in the feeding experiment had previously been studied during at least one of five breeding seasons ( and 1997). Laying dates of the first nests in a given year varied between 15 April and 23 May, and colony averages ranged from 26 April to 11 May. Mean laying dates varied significantly between years (F 29.7, df 4, 282, P ). However, using relative laying date (i.e., individual laying date minus the average of the year) to control for the effect of the year, we found no significant differences between colonies in their mean laying dates (F 1.3, df 3, 283, P 0.26; Fig. 2). Date of laying was negatively related to clutch size (Pearson correlation: r 0.37, N 257, P ), and positively related to nestling mortality (r 0.24, N 117 broods, P 0.008). The number of young fledged was lower for late-breeding pairs. Nevertheless, body condition of fledglings, measured as their pectoral muscle thickness (see Aparicio and Cordero 2001), did not depend on their hatching dates (r 0.05, N 84, P 0.65). The feeding experiment No differences in the age of kestrels were found between colony treatments (F 0.88, df 2, 54, P

4 876 NOTES Ecology, Vol. 83, No for males; F 1.19, df 2, 70, P 0.32 for females), or between fed and unfed kestrels of mixedcolonies treatments (F 0.72, df 1, 19, P 0.41 for males; F 0.73, df 1, 27, P 0.40 for females). Within mixed colonies, experimental nests were occupied earlier than control nests (F 5.60, df 1, 47, P 0.02). However, there was no difference in date of occupation between colony treatments (F 0.53, df 2, 123, P 0.59). As in previous years, no differences in laying dates were found between colonies within treatments (allunfed colonies, F 0.23, df 2, 45, P 0.79; unfed pairs of mixed colonies, F 0.63, df 1, 22, P 0.43; fed pairs of mixed colonies, F 0.34, df 1, 23, P 0.56; all-fed colonies, F 1.3, df 1, 27, P 0.26). In accordance with both hypotheses, we found that the mean laying date of fed pairs was significantly earlier than that of unfed pairs in mixed colonies (F 9.1, df 1, 47, P 0.004; Fig. 3). In these colonies, experimental feeding advanced laying date by 5.8 d on average. To test whether these differences were due to ordered nest selection occurring in mixed colonies, we compared laying dates of fed and unfed pairs of mixed colonies to those of pairs in control colonies (i.e., allfed colonies and all-unfed colonies, respectively). We found no differences in mean laying dates between fed pairs in mixed colonies and pairs in all-fed colonies (F 0.12, df 1, 52, P 0.73), or between unfed pairs in mixed colonies and pairs in all-unfed colonies (F 0.14, df 1, 70, P 0.70). Moreover, pairs in all-fed colonies started breeding significantly earlier than those in all-unfed colonies (F 8.9, df 1, 75, P 0.004; Fig. 3). The mean advance of laying date in this last comparison was 5.7 d, which was comparable to that observed in mixed colonies. DISCUSSION In mixed colonies, fed females laid earlier than unfed ones. These results are in accordance with both the FIG. 2. Laying dates (mean 1 SD) recorded at different Lesser Kestrel colonies in Spain with 10 breeding pairs during several years of study. Each symbol type refers to the same colony in different years. Laying dates refer to Julian date (day 1 1 January). Numbers indicate sample size. FIG. 3. Laying date (mean 1 SE) in each colony in relation to its treatment (all-unfed, mixed, and all-fed colonies). Squares and circles show values for fed and unfed kestrels, respectively. Julian dates (day 1 1 January) are used. Numbers indicate sample size. Habitat Selection and the Food Supply Hypotheses. However, contrary to what is predicted by the Habitat Selection Hypothesis, the mean laying dates of unfed and fed pairs within mixed colonies did not differ from those of their controls (all-unfed and all-fed colonies, respectively), in which ordered nest selection was precluded by application of uniform feeding treatment. The effects of individual quality on laying date predicted by the Habitat Selection Hypothesis might be masked if, in mixed colonies, unfed pairs had taken food from food-supplemented nests. That yellow pellets containing remains of chickens were always found at or near supplemented nests precludes this possibility, because it is very improbable that unfed pairs were consuming chickens and regurgitating pellets far from their own nest. The results support the predictions of the Food Supply Hypothesis, because pairs in all-fed colonies laid significantly earlier than those in all-unfed ones. Furthermore, differences in laying dates between fed and unfed pairs in mixed colonies can be attributed to the direct effect of extra food, because the difference was the same as that between all-fed colonies and all-unfed colonies (5.8 d vs. 5.7 d, respectively; Fig. 3). Kelly and Van Horne (1997) compared years with supplemental feeding to years without feeding within a population, and they did not find significant differences in laying date. We think that it is more appropriate to compare populations within years, because timing of breeding may be more variable between years than between neighboring populations within the same year. In our study area, laying dates differed significantly between years, but not between colonies. The mean laying date of unfed birds in 1997 was as early as that of fed birds in If we used 1997 as a control, we would conclude that supplemental feeding had no

5 March 2002 NOTES 877 effect on the timing of breeding. However, such an analysis does not control for interannual changes in environmental factors that may affect breeding time (Perrins 1991, Nager and van Noordwijk 1995, Korpimäki and Wiehn 1998). The winter of 1997 was so benign that some kestrels did not emigrate, and clutches were the earliest and the largest recorded to date. Thus, natural feeding conditions in 1997 were probably as good as those created by the experimental feeding in In conclusion, we found no evidence to support the Habitat Selection Hypothesis. The effect of extra food in mixed colonies was similar to its effect in spatial reference populations (all-fed vs. all-unfed colonies) in which all members of the colony received the same treatment and there was no possibility of an ordered nest selection. The results of our experiment indicate that food supply affected laying date in a direct way, improving female ability to lay early, as suggested by the Food Supply Hypothesis. ACKNOWLEDGMENTS This study was partially supported by La Junta de Comunidades de Castilla La Mancha (projects CR01/99 and CR02/99). Chicks for the experiment were provided by Grupo de Recuperación de Fauna Autóctona (GREFA). J. M. Aparicio was under contract to the Ministerio de Educación y Cultura ascribed to the project PB R. Bonal thanks José P. Veiga for facilities during his stay at the Museo Nacional de Ciencias Naturales (MNCN). We are grateful to J. Kelly, J. R. Walters, and an anonymous referee for their valuable comments. LITERATURE CITED Aparicio, J. M The seasonal decline in clutch size: an experiment with supplementary food in the kestrel, Falco tinnunculus. Oikos 71: Aparicio, J. M Cost and benefits of surplus offspring in the lesser kestrel (Falco naumanni). Behavioral Ecology and Sociobiology 41: Aparicio, J. M Individual optimization may explain differences in breeding time in the European kestrel Falco tinnunculus. Journal of Avian Biology 29: Aparicio, J. M., and P. J. Cordero The effects of the minimum threshold condition for breeding on offspring sex-ratio adjustment in the Lesser Kestrel. Evolution 55: Arcese, P., and J. N. M. Smith Effects of population density and supplemental food on reproduction in Song Sparrows. Journal of Animal Ecology 57: Boutin, S Food supplementation experiments with terrestrial vertebrates: patterns, problems, and the future. Canadian Journal of Zoology 68: Brinkhof, M. W. G., A. J. Cave, and A. C. Perdeck The seasonal decline in the first-year survival of juvenile coots: an experimental approach. Journal of Animal Ecology 66: Cavers, P. B., and M. G. Steel Patterns of change in seed weight over time on individual plants. American Naturalist 123: Daan, S., C. Dijkstra, R. Drent, and T. Meijer Food supply and the annual timing of avian reproduction. Acta International Ornithological Congress 19: Dobson, F. S., and P. Myers The seasonal decline in the litter size of meadow voles. Journal of Mammalogy 70: Donázar, J. A., J. J. Negro, and F. Hiraldo Functional analysis of mate-feeding in the Lesser Kestrel Falco naumanni. Ornis Scandinavica 23: Franco, A., and J. Andrada Alimentación y selección de presa en Falco naumanni. Ardeola 23: Harris, M. P., D. J. Halley, and S. Wanless The postfledging survival of young guillemots Uria aalge in relation to hatching date and growth. Ibis 134: Kelly, J. F., and B. Van Horne Effects of food supplementation on the timing of nest initiation in Belted Kingfishers. Ecology 78: Koenig, W. D., and S. S. Albano Lifetime reproductive success, selection, and the opportunity for selection in the white tailed skimmer, Plathemis lydia (Odonata: Asioptera). Evolution 41: Korpimäkki, E., and J. Wiehn Clutch size of kestrels: seasonal decline and experimental evidence for food limitation under fluctuating food conditions. Oikos 83: Landa, K Seasonal declines in offspring fitness and selection for early reproduction in nymph-overwintering grasshoppers. Evolution 46: Martin, T. E Food as a limit on breeding birds: a lifehistory perspective. Annual Review of Ecology and Systematics 18: Nager, R. R., and A. J. van Noordwijk Proximate and ultimate aspects of phenotypic plasticity in timing of great tit breeding in a heterogeneous environment. American Naturalist 146: Negro, J. J., F. Hiraldo, and J. A. Donázar Causes of natal dispersal in the lesser kestrel: inbreeding avoidance or resource competition? Journal of Animal Ecology 66: Olsson, M., and R. Shine The seasonal timing of oviposition in sand lizards (Lacerta agilis): why early clutches are better. Journal of Evolutionary Biology 10: Perrins, C. M The timing of birds breeding seasons. Ibis 112: Perrins, C. M Tits and their caterpillar food supply. Ibis 33(Supplement): Schultz, E. T The effect of birth date on fitness of female dwarf perch, Micronometrus minimus (Perciformes: Embiotocidae). Evolution 47: Schultz, E. T., L. M. Clifton, and R. R. Warner Energetic constraints and size-based tactics: the adaptive significance of breeding-schedule variation in a marine fish (Embiotocidae: Micronometrus minimus). American Naturalist 138: Svensson, E Avian reproductive timing: when should parents be prudent? Animal Behaviour 49: Ydenberg, R. C., C. W. Clark, and A. Harfenist Intraspecific fledging mass variation in the Alcidae, with special reference to the seasonal fledging mass decline. American Naturalist 145: