Late 19th-Century Navigational Uncertainties and Their Influence on Sea Surface Temperature Estimates

• Activity Number: 227 - The Best of Annals of Applied Statistics
• Type: Invited
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: IMS
• Author(s): Chenguang Dai* and Duo Chan and Peter Huybers and Natesh Pillai
• Companies: Harvard University and Harvard University and Harvard University and Harvard University
• Keywords: state-space model; hierarchical model; position error; navigational uncertainty; sea surface temperature uncertainty

Abstract

Accurate estimates of historical changes in sea surface temperatures (SSTs) are important. A source of uncertainty that has not been quantified in historical SST estimates stems from position errors. A Bayesian inference framework is proposed for quantifying errors in reported positions and their implications on SST estimates. The analysis framework is applied to data from ICOADS3.0 in 1885. Focus is upon a subset of 943 ship tracks, of which the positions are determined by dead reckoning that are periodically updated by celestial correction techniques. The posterior medians of uncertainties in celestial correction are 0.30 degree in longitude and 0.22 degree in latitude, respectively. The posterior medians for two-hourly dead reckoning uncertainties are 19.2% for ship speed and 13.2 degree for ship heading. We then translate position errors into SST uncertainties by sampling an ensemble of SSTs from the MURSST data set. Evolving technology for determining ship position, heterogeneous archiving of position information, and seasonal and spatial changes in SST gradients together imply that accounting for positional error in SST estimates will require substantial additional effort.