Landscape characteristics of successful breeding colonies of the double-crested cormorant

Background/Question/Methods

Widespread use of persistent organic pesticides in the 20^th century influenced the population size and distribution of many North American predatory species. The Double-crested Cormorant (Phalacrocorax auritus, DCCO), experienced a population crash during this period that brought their once thriving population of millions of birds to under ten thousand individuals. Today the population is considered to be recovered after protection of the species through local and national legislations and the discontinuance of DDT in North America. The recovery of the DCCO took decades and multiple generations of people lived in their historical breeding and wintering range with few sightings of this species. Consequently, the recovery of this piscivorous bird has sparked human-wildlife conflicts as in some areas (within their traditional range of distribution) they are considered to be an invasive nuisance species. Lethal management of DCCO is underway in a dozen states and is proposed in many more as the new range of this predator exceeds that of the historical population. DCCO are now found breeding and wintering in locations where suitable habitat were not previously available, but after anthropogenic landscape alteration, are now thriving. We used a species distribution model (MaxEnt) based on documented distributions of cormorants and their common prey species (fishes), geological and climactic data, and human-induced alteration to the landscape and biome of Minnesota and Florida. 

Results/Conclusions

Density of nearby (within 3,500m) water availability, fish stocking parameters, availability of island and swamp habitat, and fish advisories nearby were the major contributors to the most explanatory model. The output maps closely followed existing colonies and were consistent in predicting large, long-term colonies. The models for Minnesota and Florida were not consistent; suggesting that birds nesting in different parts of the country may cue in on different landscape features and habitat types. This information can be used to estimate the potential carrying capacity of even larger regions where the distribution of DCCO is spreading because of continued landscape alteration.