Combining regional habitat selection models for large-scale prediction : circumpolar habitat selection of southern ocean humpback whales

• Published in: Remote Sensing
• Main Authors: Reisinger, Ryan Rudolf, Friedlaender, Ari S., Zerbini, Alexandre N., Palacios, Daniel M., Andrews-Goff, Virginia, Rosa, Luciano Dalla, Double, Mike, Findlay, Kenneth Pierce, Garrigue, Claire, How, Jason, Jenner, Curt, Jenner, Micheline-Nicole, Mate, Bruce, Rosenbaum, Howard C., Seakamela, S. Mduduzi, Constantine, Rochelle
• Format: Article in Journal/Newspaper
• Language: English
• Published: MDPI 2021
• Subjects: Ensembles; Habitat selection; Machine learning; Prediction; Resource selection functions; Telemetry; Megaptera novaeangliae; Humpback whale (Megaptera novaeangliae); Southern Ocean; Humpback Whale
• Online Access: https://doi.org/10.3390/rs13112074; https://repository.up.ac.za/handle/2263/88126

SUPPLEMENTARY MATERIALS : Table S1: Tracking data. Table summarizing the tracking data collated for this study. Supplementary Figure S1: Partial dependence plots. Relationship between the regional model predictions and the four most important environmental covariates, in order of decreasing mean importance: ICEDIST, SLOPEDIST, SST, and SHELFDIST. Partial dependence plots show the predicted response probability, here p(Observed track), on the vertical axis, over values of the environmental covariate in question while accounting for the average effect of the other predictors in the model [114]. DATA AVAILABILITY STATEMENT : Computer code and derived data are in the paper’s Github repository: https://github.com/ryanreisinger/megaPrediction (accessed on 21 May 2021) Machine learning algorithms are often used to model and predict animal habitat selection— the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar ...