Presentation: Modelling Ocean Temperatures from Bio-probes Under Preferential Sampling
Speaker: Matias Salibian-Barrera, Ph.D., Professor of Statistics, The University of British Columbia
Abstract: In the last 25 years there has been an important increase in the amount of data collected from animal-mounted sensors (bio-probes) which are often used to study the animals’ behaviour or environment. We focus here on the latter, where the interest is in sea surface temperature (SST), and measurements are taken from sensors mounted on elephant seals in the Southern Indian Ocean. We show that standard geostatistical models may not be reliable for this type of data, due the potential presence of preferential sampling, which occurs when the locations at which observations are made may depend on the spatial process that underlines the correlation structure of the variable of interest. Ignoring this phenomenon may affect the resulting spatial predictions and parameter estimates. We first show that previously proposed Monte Carlo estimates for the likelihood function in these models may not be reliable and that in the case of stationary sampling locations widely available numerical methods can be used to approximate the likelihood function. Finally we extend this methodology to observations obtained by devices that move through the region of interest, as is the case with the tagged seals. Our simulation studies confirm that predictions obtained from the preferential sampling model are more reliable when this phenomenon is present, and that they compare very well to the standard ones when there is no preferential sampling.
This is joint work with my former PhD student, Dr. Daniel Dinsdale.