Territory location and quality, together with climate, affect the timing of breeding in the white-throated dipper

Abstract

Recent climate change has led to advanced spring phenology in many temperate regions. The phenological response to variation in the local environment, such as the habitat characteristics of the territories birds occupy, is less clear. The aim of this study is to understand how ecological conditions affect breeding time, and its consequences for reproduction, in a white-throated dipper Cinclus cinclus population in a river system in Norway during 34 years (1978-2011). Hatching date advanced almost nine days, indicating a response to higher temperatures and the advanced phenology in the area. Earlier breeding was found in warm springs and at lower altitudes. High population density facilitated earlier breeding close to the coast. Furthermore, when population density was low, breeding was early at territories that were rarely occupied, while in years with high density, breeding was early at territories that were frequently occupied. Also, when population density was low, earlier breeding occurred at territories that on average produced more offspring than other territories, while there was no difference in breeding time in high population years. Selection for early breeding was dependent on spring temperatures and high spring temperatures contributed to higher breeding success during the study period. We found that breeding phenology may have strong effects on fitness in the white-throated dipper, and thus that breeding time is an important ecological factor in a species that feeds mainly on aquatic rather than terrestrial prey.

Figure 1

The annual variation in hatching date and spring temperatures. (a) Annual variation in hatching date (hatching day-of-year, solid line) and mean spring temperatures (Feb–Apr, broken line) in the white-throated dipper population in Lyngdalselva, and (b) the effect of mean spring temperature on annual hatching date in the study area (1978–2011). The trend line in each time series is denoted with a solid line. N = 1045.

Figure 2

The causes of variation in the timing of breeding modelled in a linear mixed-effects model in the white-throated dipper in Lyngdalsvassdraget 1978–2011. Effects are shown for the following fixed effects, using model predictions and a restricted data set: (a) altitude (m above sea level), (b) the interaction between distance from the coast (m) and population density, where population density is ordered from low to high with decreasingly light grey tones, (c) the interaction between overall mean territory occupancy rate and population density, where population density is ordered as in (b), and (d) the interaction between overall mean territory brood size and population density, where population density is ordered as in (b) and (c),with the crossed random effects year, female identity and territory identity, on the hatching day-of-year. N = 992.

Figure 3

The consequences of variation in the timing of breeding, measured as the number of fledglings given the number of eggs laid in the white-throated dipper in Lyngdalsvassdraget 1978–2011. The effects are shown for the interactions (a) between spring temperature and hatching day-of-year, where spring temperature is ordered from low to high with increasingly light grey tones where −1.5–−0.5 is symbolised by solid line and circles, −0.5–0.5 °C by broken line and triangles, 0.5–1.5 °C by dotted line and plus symbols, 1.5–2.5 °C by broken-dotted line and crosses, 2.5–3.5 °C by long-stapled line and diamonds, and 3.5–4.5 °C by long-stapled-dotted line and downwards-facing grey triangles, and (b) between spring temperature and year, where spring temperature is ordered from the earliest to the latest as in (a), from a mixed-effects model including the most influential predictor variables (RI > 0.5) with the random crossed effects female and territory identity. N = 999.

Figure 4

The consequences of variation in the timing of breeding, measured as the number of recruits given the number of fledglings, depending on the breeding time, measured as the hatching day-of-year, in the white-throated dipper in Lyngdalsvassdraget 1978–2011. N = 999.