Integrating Ocean Observing Data to Model Marine Animal Distributions

Patrick N. Halpin Marine Geospatial Ecology Lab, Duke University, USA

phalpin@duke.edu 

Forecasting the distribution and density of marine animals requires accurate characterization of dynamic oceanographic habitats. The characterization of these habitats often relies on development of environmental covariates from remotely sensed earth observing data. The development of environmental covariates raises many considerations concerning the appropriate oceanographic indices to be developed and the spatial and temporal scales of assessment. These considerations include the appropriate use of contemporaneous versus climatological covariates in the development of explanatory models. This presentation will discuss this topic using a broad range of cases as well as a focus on the choices and implications considered in the development of environmental covariates for the prediction of cetacean species along the US Atlantic coast. Many highly migratory marine animals in this region are vulnerable to human interactions such as ship strikes, fishing gear entanglement, and ocean noise habitat disruption. We combine times series of remotely sensed earth observing data with ship and aircraft observer data to create temporally dynamic forecasting models of species distributions and densities. In this presentation we depict the process from data collection, data aggregation, oceanographic feature development to forecast modeling.

Recent publications:

LaBrecque, Curtice, Harrison, Van Parijs, Halpin (2015) Biologically Important Areas for Cetaceans Within US Waters–East Coast Region Aquatic Mammals 41, 17-29

P Tepsich, M Rosso, PN Halpin, A Moulins (2014) Habitat preferences of two deep-diving cetacean species in the northern Ligurian Sea. Mar. Ecol. Prog. Ser. 508, 247-260

Contact:  Ana Rodrigues : ana.rodrigues@cefe.cnrs.fr 

Contact du Comité SEEM : seem@services.cnrs.fr.   

Contact du Labex CEMEB: gestion.cemeb@univ-montp2.fr,  www.labex-cemeb.org.