The ecological niche of storm-petrels in the North Pacific and a global model of dimethylsulfide concentration

Abstract

Ecological niche modeling techniques were used to create global, monthly predictions of sea surface dimethylsulfide (DMS) concentrations, and breeding season distribution of Leach's Storm-Petrel (Oceanodroma luncorhoa) and Fork-Tailed Storm-Petrel (O. furcata) in the North Pacific. This work represents the first attempt to model DMS concentrations on a global scale using ecological niche modeling, and the first models of Storm-Petrel distribution for the North Pacific. Storm-Petrels have been shown to be attracted to DMS, and it is therefore likely that a model of sea surface DMS concentration would help explain and predict Storm-Petrel distribution. We have successfully created the most accurate models of sea surface DMS concentrations that we are currently aware of with global correlation (r) values greater than 0.45. We also created Storm-Petrel models with area under the receiver operating characteristic curve (AUC) values of greater than 0.90. Using just DMS as a predictor variable we were also able to create models with AUC values upwards of 0.84. Future conservation efforts on pelagic seabird species may be dependent on models like the ones created here, and it is therefore important that these methods are improved upon to help seabird management on all scales (global, national, regional and local).