Portal de Periódicos da CAPES

Food, Predation, and Reproductive Ecology of the Dark-Eyed Junco in Northern Utah

1982; Oxford University Press; Volume: 99; Issue: 4 Linguagem: Inglês

1938-4254

Kimberly G. Smith, Douglas C. Andersen,

Animal Behavior and Reproduction

-During 1976, 1977, and 1978, 26, 29, and 19 nests, respectively, of the groundnesting, Dark-eyed Junco (Junco hyemalis mearnsi) were monitored in a sprucefir ecosystem in northern Utah. In 1976 and 1977, 4-egg clutches, found mainly in June through mid-July, produced more young than did 3-egg clutches, found after mid-July. In these two years, nestlings from 4-egg clutches in which all eggs hatched gained significantly more weight than did nestlings from 3-egg clutches. In 1978, only 5and 4-egg clutches were found, and these nestlings were intermediate in weight between nestlings from the 4and 3-egg clutches of the previous years but not significantly different from nestlings of either clutch size. Growth rates (K) of body weight were not significantly different among nestlings from all clutch sizes, so that the rate of weight gain seems independent of clutch size in the Pink-sided Junco. Final tarsal length upon leaving the nest also did not differ significantly among clutch sizes, but the rate of tarsal growth did. The length of tarsi was independent of clutch size, and the length upon leaving the nest was equal to that of adult juncos. We interpret this as support for the hypothesis, as tarsal length should have high functional importance in passerines that run rather than fly from the nest. Populations of major food items and potential nest predators were monitored. Although climatic conditions varied dramatically during the three seasons we studied these juncos, the dry biomass of food items in junco diets peaked 1 week after the mean hatching date of 4-egg clutches in all three years. We suspect that nestlings from late 3-egg clutches experienced reduced food availability, which explains the difference in weight between nestlings from 3and 4-egg clutches. Total rodent predation pressure remained at or above 10 individuals/ha, although populations of the species involved fluctuated between years. Increased predation was noted in 1977, when weasel populations were unusually high. Latespring snowstorms and late-summer thunderstorms also probably influence nesting success. We hypothesize that predation pressure and short periods of both food abundance and favorable weather influence the initiation of breeding, clutch size, and nestling-growth patterns and lead to a decline in clutch size as the breeding season progresses. Received 14 November 1979, resubmitted 9 November 1981, accepted 11 April 1982. NUMEROUS models have been proposed to account for variation ifi avian clutch size in terms of both proximate and ultimate factors (reviewed by Klomp 1970). Lack's (1954: 22) suggestion that food supply directly affects mean clutch size has been supported by field observations (Bryant 1975 and references therein), natural experiments (Anderson 1977), and controlled field experiments (Hussell 1972, 1 Present Address: Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72701 USA. 2 Present Address: Department of Forestry and Natural Resources, Purdue University, West Lafayette, Indiana 47907 USA. Crossner 1977). In a recent review of experimental evidence, however, Hogstedt (1981) concluded that factors other than food must also be important. Such factors would include predation, competition, and territoriality (e.g. Cody 1966, Brockelman 1975, Dhondt 1977, Perrins 1977, Hogstedt 1980). Undoubtedly, many factors act in concert to produce variation in clutch size within and among species; hence, simple explanations encompassing only one factor should not be expected to explain variation in clutch size over broad areas or among diverse avian taxa. Ultimately, natural selection should promote the clutch size that results in the greatest parental fitness (e.g. Charnov and Krebs 1974, Perrins and Moss 1975; but see De Steven 1980). 650 The Auk 99: 650-661. October 1982 This content downloaded from 157.55.39.107 on Wed, 30 Mar 2016 05:33:42 UTC All use subject to http://about.jstor.org/terms October 1982] Junco Breeding Ecology 651 Research on avian reproductive ecology has expanded to include an examination of selective factors affecting growth and development of nestlings (Ricklefs 1967, 1968a, 1976; Austin and Ricklefs 1977; O'Connor 1977a, and references therein). The adaptive-growth states that a nestling's resources are allocated to the growth of those body components with the highest current functional priority but also with regard to future needs (for review see O'Connor 1977a). Because functional priorities (e.g. development of flight muscles, feathers, or digestive organs) are different among species having different nesting habits, the adaptivegrowth hypothesis predicts differences in the growth patterns of different species (e.g. Austin and Ricklefs 1977, O'Connor 1977a). In this report, we examine the breeding ecology of the Pink-sided Junco (Junco hyemalis mearnsi), a member of the Dark-eyed Junco complex. We correlate several direct determinants of fitness (clutch size, nesting success, and growth rates) with environmental features (available food supply, predator populations, and weather conditions) and suggest causal relationships among them. We contend that selection has operated to produce a suite of interdependent characteristics allowing the Pinksided Junco to occupy a relatively harsh and unpredictable environment, and, based on our observations, we suggest an adaptive strategy, which we present as hypotheses to be tested by other independent studies. STUDY AREA AND METHODS The Pink-sided Junco is the most common groundnesting, ground-feeding, insectivorous bird found above 2,000 m in northern Utah. Females were flushed from nests incidental to extensive foot travel by ourselves and co-workers throughout about 2 km2 of mixed subalpine meadows, aspen (Populus tremuloides) forest, and spruce-fir (Picea engelmannii-Abies lasiocarpa) forest in or near the Utah State University School Forest (elevation = 2,600 m), Cache and Rich counties, Utah (41?52'N, 111?30'W). Although no systematic search for nests was made, most of the area was repeatably visited by us and cooperating workers advised in our techniques for locating nests. A more thorough description of the study site is presented in Schimpf et al. (1980). Data were collected during the summers of 1976, 1977, and 1978. Upon discovery, nests were visited almost daily, but no set visitation time was followed. Nestlings were weighed to the nearest 0.25 g using a 30-g Pesola spring scale. In 1977 and 1978, the right tarsus was measured to the nearest 0.1 mm with dial calipers, and individual nestlings were identified by marking tarsi with indelible black ink. Average weighing times of nestlings ranged between 1200 and 1600, and most were between 1300 and 1500. Nestlings were banded with USFWS aluminum bands on about Day 8; a few chicks banded at earlier ages were removed from the nest by parents. Nest losses due to banding were not included in the analysis. Weights and measurements of adult and juvenile juncos were obtained from birds mist-netted on the study site during summers of 1976 and 1977. Females that laid 4-egg clutches initiated incubation after the 3rd egg was laid. The first 3 eggs hatched synchronously 11-12 days after incubation was initiated, while the 4th egg hatched later the same day or the next morning. For each individual nestling, the day of hatching was considered Day 0 in all analyses, so that in some nests one nestling was a day younger than its siblings (two nestlings in a few fiveegg nests). Eggs hatched more or less synchronously in three-egg clutches. Only nests in which all eggs hatched are included in the analyses of weight gain and tarsi growth. Of the 73 nestlings used in the analysis of weight gain, 44 were first measured on Day 0 or Day 1, 21 on Day 2, and 8 on Day 3. Based on our observations, we assumed in the analysis that all nestlings weighed 1.75 g on Day 0 if they were not measured on that day. We analyzed the increase in body weight and length of tarsus by fitting logistic growth curves to the data points of individual nestlings (see Ricklefs 1979). The logistic curve has the form: M(t) = A/[1 + exp(-K[t ti])] where M(t) is the weight (g) at age t (days), A is the asymptote (g) of the growth curve, K is the growth rate constant (days-'), and ti is the age at the inflection point. The curves were fitted using the nonlinear least-squares regression procedure (NLIN) available in the SAS statistical package. Only individuals with four or more data points were included in the analysis. Also, individuals were deleted from further analysis if the standard error of the asymptotic estimate was greater than 10% of the estimated asymptote. Differences in asymptotes (final weights) and rates of growth (K) among clutch sizes and years were then examined using t-tests. The stomach contents of three adult juncos shot while actively feeding within the study area (1 each on 20 June, 13 July, and 21 July 1977) were examined to determine major food items. The stomach-content data corroborated results of other studies (see below) and supported our field observations that seeds are not important in junco diets during breeding on our study site. Estimates of abundance and biomass of known and suspected food items (arthropods) were made available to us by James A. MacMahon (Utah State University). These data were collected concurrently with ours in two meadows within which several nests were found. Random locations within This content downloaded from 157.55.39.107 on Wed, 30 Mar 2016 05:33:42 UTC All use subject to http://about.jstor.org/terms 652 SMITH AND ANDERSEN [Auk, Vol. 99 ~~~~~~~~~~~~~~~ .......