An Integrated Ocean-Atmosphere Approach to Reconstructing the Past and Present Climate
Re-analysis products, also known as data assimilation or state estimation products, are combinations of observations and numerical models that offer the best reconstruction of what actually happened in the climate system, overcoming issues due to the sparsity of ocean observations through the past century. Atmosphere-only re-analyses have proven invaluable for scientific study, but they contain global imbalances that can obscure climate trends. Some ocean-only re-analyses, headlined by the ECCO (Estimating the Climate and Circulation of the Ocean) suite of products, are designed to exactly follow the equations of motion in a state-of-the-art numerical model, which allows ocean phenomena to be attributed to physical mechanisms. While the ECCO Consortium has made a major step forward, their reconstructions of winds and air-sea heat and freshwater exchange are contaminated by numerical model errors, which has slowed the adoption of the re-analysis product by the wider community interested in oceans and climate.
Conceptually, these problems can be resolved by treating the ocean and atmosphere as a coupled system where climate-relevant quantities are tracked and balanced across the air-sea interface. To develop a coupled ocean-atmosphere re-analysis that is faithful to the laws of physics, we propose to extend the ECCO ocean re-analysis system for use with a coupled ocean-atmosphere numerical model. Atmospheric weather, however, has caused problems for any re-analysis longer than a week, and a proof-of-concept is necessary to show that these problems can be overcome. Here we propose to investigate the adequacy of simplified atmospheric models, such as a linear empirical model or an energy balance model, for capturing ocean-atmosphere interaction on climate timescales, while removing the problems associated with reconstructing individual atmospheric storms. The successful proof-of-concept will accurately reconstruct the known modern-day circulation with fewer inaccuracies at the sea surface, which will lead to greater confidence in the ocean reconstruction and greater adoption by the WHOI and wider oceanographic communities.
Historically, atmospheric and oceanic re-analyses have been developed mostly independently, and the goal of this project is to leverage and merge the two lines of development. A promising further step is to apply the ocean-atmosphere re-analysis method to climate records to diagnose the cause of decadal and longer-term variability, whether it originates in the ocean, atmosphere, or external to the system. In addition, the ocean-atmosphere re-analysis would also provide an improved way to initialize future climate forecasts and to reconstruct very different climate states such as those during the last Ice Age.