Sexual size dimorphism and sex determination by morphometric measurements in breeding Imperial Shags (Phalacrocorax atriceps)

• Main Authors: Svagelj, Walter Sergio, Quintana, Flavio Roberto
• Format: Article in Journal/Newspaper
• Language: English
• Published: Waterbird Society
• Subjects: DISCRIMINANT ANALYSIS; IMPERIAL SHAGS; PHALACROCORACIDAE; PHALACROCORAX ATRICEPS; SEXUAL SIZE DIMORPHISM; https://purl.org/becyt/ford/1.6; https://purl.org/becyt/ford/1; Patagonia; Argentina; Imperial Shag; Phalacrocorax atriceps
• Online Access: http://hdl.handle.net/11336/102934

The Imperial Shag (Phalacrocorax atriceps) is monomorphic in plumage, but males are larger than females. We analyzed the sexual size dimorphism and variability of six morphometric characteristics (bill length, bill depth, head length, tarsus length, wing length, and body mass) measured on 291 breeding Imperial Shags at Punta León colony in coastal Patagonia, Argentina, during 2004 and 2005 breeding seasons. Discriminant analyses were performed on external measurements that we considered potentially useful in sexing Imperial Shags. All the birds were sexed by a distinctive behavior (vocalizations) and a sub-sample of fifty were also sexed by DNA-based genetic techniques, showing 100% agreement between the two methods. All measured characteristics differed between the sexes, with males being larger than females. Body mass (17.8 - 18.0%) and bill depth (12.8 - 13.2%) showed the highest level of dimorphism whereas bill, head, tarsus, and wing length were less dimorphic (ranging 4.8 - 6.0%). Dimorphism in body mass for breeding shags was lower than previously reported during winter, prior to the breeding season. Although the wing length showed the lowest degree of dimorphism (5.3 - 5.4%), it presented the lowest coefficient of variation (1.9 - 2.0%) resulting in the most accurate single-measurement indicating sex. A cross validation process with a new sample, revealed that discriminant functions comprised by two characteristics were more accurate and reliable for sex determination than single-measurements. We obtained two functions correctly classifying 94 - 97% of shags. These functions were reliable (similar accuracy for discriminant analysis, Jackknifed validation, and cross-validation with a new sample) and seasonal unbiased, as body mass was not included in the analysis. Fil: Svagelj, Walter Sergio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Nacional Patagónico; Argentina Fil: Quintana, Flavio Roberto. Wildlife Conservation Society; Estados Unidos. Consejo Nacional de Investigaciones ...