Distribution of common dolphins (Delphinus spp.) in the western Atlantic Ocean: a critical re-examination

Abstract

Due to indications that misidentification (largely confusion among dolphins of the genera Delphinus and Stenella) in the past had led to erroneous assumptions of distribution of the two species of common dolphins (Delphinus delphis and D. capensis) in the western Atlantic Ocean, we conducted a critical re-examination of records of the genus Delphinus from this region. We compiled 460 ‘plottable’ records, required support for confirmation of genus and species identifications, and found many records lacking (and some clearly misidentified). When we plotted only the valid records (n = 364), we found evidence of populations in only three areas, and apparent absence throughout much of the tropical/subtropical regions. Off the east coast of the US and Canada, D. delphis is found from the Georgia/South Carolina border (32°N) north to about 47–50°N off Newfoundland. Since the 1960s, they have apparently been absent from Florida waters. There is no evidence that dolphins of the genus occur in the Gulf of Mexico. Reports of common dolphins from most of the Caribbean Basin are also rejected, and the only place in that region where they are confirmed to occur is off central-eastern Venezuela (a coastal D. capensis population). Off eastern South America, common dolphins appear to be restricted to south of 20°S. There is a coastal long-beaked population found in the South Brazil Bight, and one or more short-beaked populations south and offshore of this (ranging south to at least northern Argentina). The results are very different from commonly-accepted patterns of distribution for the genus in the Atlantic. Most areas of distribution coincide with moderate to strong upwelling and common dolphins appear to avoid warm, tropical waters. This study shows that great care must be taken in identification of similar-appearing long-beaked delphinids, and that uncritical acceptance of records at face value can lead to incorrect assumptions about the ranges of the species involved.