The black-headed duck (Heteronetta atricapilla) of South America is the only known avian obligate brood parasite with precocial offspring. In Argentina, it relies on two species of coots as primary hosts, which typically reject 35-65% of duck eggs. We show that environmentally-driven increases in host egg-rejection behavior lead to substantial reductions in the reproductive success of the brood parasite. Episodes of flooding and vegetation loss caused dramatic shifts in host egg rejection behavior, resulting in rejection (85-95%) of almost all duck eggs. Coots respond to fluctuating water levels by building up their nest, raising their own eggs but leaving duck eggs behind. Coots can apparently recognize parasitic duck eggs, but large-scale rejection is triggered only when hosts must actively make a choice. We use a simple population model to illustrate the unique demographic challenges that black-headed ducks face with their parasitic lifestyle, and to explore the potential impact of environmentally-induced escalation of egg rejection. Using best available estimates for key vital rates, we show that obligate parasitism may provide a demographically precarious existence for black-headed ducks, even under benign environmental conditions. Environmentally-mediated increases in egg rejection rates by hosts could impact significantly the viability of this enigmatic species of brood parasitic duck. Our results demonstrate that egg rejection rates are not fixed properties of host populations or individuals but are strongly influenced by social and ecological factors. Shifts in these environmental drivers could have important and unforeseen demographic consequences for brood parasites.

Field observations and experiements in Argentina in 1993, 1994 and 1997.We conducted systematic surveys of the marshes every two to four days on foot or by canoe to find potential host nests and detect brood parasitism. Nests were identified to species by observing birds on or near nests. Parasitism was easily detected because the duck eggs differ dramatically from the eggs of both major hosts. Environmental effects on egg rejection were determined using four methods: natural cases of brood parasitism by the ducks (Mal Abrigo Gull and Tern Marshes and Real Viejo Marsh A, all 1993), experimental parasitism with real duck eggs swapped among host nests as soon as the eggs were laid (Real Viejo Marsh B 1993), experimental parasitism with painted domestic chicken eggs whose length and width (length: 50.5-63.5 mm, width: 41.3-47.6 mm, n = 314 eggs) overlapped the size range of duck eggs (Cari Lauquen 1994, Real Viejo Marshes A and B 1997), and experimental parasitism with real coot eggs from nests of conspecifics (Cari Lauquen 1994, Real Viejo Marshes A). The diversity of approaches was employed originally to examine different aspects of the host-parasite interaction (e.g. Lyon and Eadie 2004, 2013), but when the environmental changes occurred, we realized that we could leverage these experiments to evaluate how the changing environmental conditions affected rejection behavior of the hosts. We used nominal logistic regression analysis to evaluate the effects of flooding, species of host, and wetland site on rejection rates of parasitic eggs in 1993.We used a Fisher’s exact test to contrast rejection rates before and after a second flood event in 1997 on one wetland (Real Viejo Marsh B) in an experimental study of egg rejection; this comparison involved a single wetland (Marsh B), one host species (red-gartered coots) and a single egg type (painted hen egg). We used nominal logistic regression analysis to evaluate the effects of vegetation loss, wetland site, and egg type on rejection rates of parasitic eggs.  For each wetland we pooled all egg rejections in that year. 

Read Me file is self explanatory.