Computer model calibration based on image warping metrics: an application for sea ice deformation

TL;DR

This paper introduces a novel calibration framework for sea ice models that uses image warping metrics to better align simulated features with observational data, addressing limitations of traditional methods.

Contribution

It develops a statistical emulation and calibration approach that incorporates image registration techniques to improve model-data alignment for features like fractures in sea ice.

Findings

Effective alignment of model outputs with observational features.

Improved calibration accuracy over traditional least-squares methods.

Applicable to other models with coherent structures.

Abstract

Arctic sea ice plays an important role in the global climate. Sea ice models governed by physical equations have been used to simulate the state of the ice including characteristics such as ice thickness, concentration, and motion. More recent models also attempt to capture features such as fractures or leads in the ice. These simulated features can be partially misaligned or misshapen when compared to observational data, whether due to numerical approximation or incomplete physics. In order to make realistic forecasts and improve understanding of the underlying processes, it is necessary to calibrate the numerical model to field data. Traditional calibration methods based on generalized least-square metrics are flawed for linear features such as sea ice cracks. We develop a statistical emulation and calibration framework that accounts for feature misalignment and misshapenness, which involves optimally aligning model output with observed features using cutting edge image registration techniques. This work can also have application to other physical models which produce coherent structures.