Effects of Food Supplementation on the Timing of Nest Initiation in Belted Kingfishers

Abstract

Numerous food supplementation experiments have tested the hypothesis that timing of reproduction in birds is limited by energy constraints. These studies have generally concluded that food availability is the primary proximate factor affecting the timing of nesting in birds. None of these studies, however, has examined the relationship between food availability and the timing of nesting in piscivorous birds. To determine whether food was the primary proximate factor affecting timing of nest initiation in the piscivorous Belted Kingfisher (Ceryle alcyon), we conducted a food supplementation experiment from 1993 through 1995; additional descriptive data were collected in 1992. Belted Kingfishers that nested early had heavier nestlings and were more likely to renest, in the event of nest failure, than were late-nesting birds. This indicates a reproductive advantage in early nesting. In 1994 and 1995, when we began the feeding treatment early in the prebreeding season (8 March) and provided enough fish to meet the estimated energy requirements of a pair of kingfishers (90 g/d), nests of supplemented birds were initiated earlier than those of unsupplemented birds as predicted by the energy constraint model. In 1993, when we began the feeding treatment later (20 April) and provided less fish (20–30 g/d), there was no difference in the initiation dates of supplemented and unsupplemented nests. Although these data support the primary prediction of the energy constraint model, two patterns in our data were not explained by this model. First, few unsupplemented nests were initiated early in the breeding season relative to an unmanipulated reference population. Second, no supplemented nests were initiated outside the range of nesting dates recorded in our unmanipulated reference population. Based on these patterns, we argue that variation in arrival time of females on the breeding grounds, coupled with their selection of territories based on food availability, offers a feasible and testable alternative to the energy constraint model. This habitat selection model predicts that food availability influences the location of reproduction of individuals, but does not alter the timing of their reproduction. A literature review revealed few data that could distinguish the energy constraint and habitat selection models. We think that it is important to consider the habitat selection alternative before concluding that the results of food supplementation studies support the energy constraint model. We suggest that future studies use spatial and temporal controls to partition variation in nest initiation dates into multiple causes (e.g., energy constraints, habitat selection).