Variation in nuptial color in relation to sex, individual quality and mating success in the sex-role reversed Phalaropus fulicarius (Red Phalarope)

Abstract

Abstract In most bird species, males are more ornamented and compete for females, who contribute more to offspring care. In a minority of species this pattern is reversed, with more colorful females competing for mates and males taking care of parental duties. In such sex-role reversed species, the links between colorful ornamentation, individual quality and mating success are not well established. Phalaropus fulicarius (Red Phalarope) is a colorful sex-role reversed migratory shorebird with regular social polyandry, in which both sexes show considerable color variation. Here, we describe sex differences in color and quantify associations between color variation and indicators of quality and mating success in both sexes. Using a large sample of photos collected across three consecutive years on the Arctic breeding grounds, we scored color variation for four body parts (bill, crown, cheek and breast), and analyzed scores separately and combined into an overall color score. Females were more colorful and larger than males, and individuals could be unambiguously sexed by crown color. Nevertheless, there was substantial variation within sexes and some overlap between males and females in bill, cheek, breast, and overall color scores. Assortative mating by color was only found for the bill. Color variation did not correlate with plasma testosterone levels, except for male cheek color. Females in better body condition had yellower bills and higher overall color scores, while early-arriving birds had higher breast and overall scores. Phalaropus fulicarius that bred locally were heavier than those that did not, but they did not have higher color scores. Female color variation did not predict the probability of local social polyandry nor variation in clutch size, and male coloration did not predict the probability of nest predation. In conclusion, P. fulicarius color variation showed modest correlations with individual quality and was unrelated to variation in local reproductive success.