Authors: Marilyn Raphael*, UCLA, Mark Stephen Handcock, UCLA Department of Statistics
Topics: Climatology and Meteorology
Keywords: Antarctic sea ice, statistical modelling
Session Type: Paper
Start / End Time: 5:20 PM / 7:00 PM
Room: Oak Alley, Sheraton, 4th Floor
Presentation File: No File Uploaded
Satellite-observed, total Antarctic sea ice extent (SIE) experiences a distinct annual cycle, peaking in September and troughing in March on average. The amplitude and phase of the annual cycle also varies regionally. What forces the observed annual cycle and its variations is not completely understood. The annual cycle may be calculated by simply taking the average SIE for each day of the year. However, while simple and transparent, this method produces a value that is subject to substantial variation since it is based on fewer than 40 numbers, one for each year of observed data. It also disguises the fact that the annual cycle might be slowly changing phase and that the amplitude as well as shape of the daily extent might vary. Here, we present a model that allows the mathematical and stochastic representation of the proximate forces that lead to the observed annual cycle of sea ice extent. These mathematical and stochastic methods allow amplitude and phase dilation and contraction. Thus, the annual cycle is not constrained to be a fixed cyclical pattern rather, it is a pattern that allows both temporal dilation and contraction as well as amplitude modulation. We use this model in an ensemble interpolation to reconstruct missing daily data in the early part of the satellite- observed sea ice data set. Results are presented and discussed.