Identifying Mechanisms of AMOC Variability in ECCO State Estimates and CMIP5 Models

Our work seeks to investigate the connection between low-frequency variability of Atlantic sea surface temperatures (SSTs) and the Atlantic Meridional Overturning Circulation (AMOC). Observations indicate Atlantic SSTs exhibit significant low-frequency variability, which impacts regional and global climate. Prior studies have linked variability of the AMOC to decadal Atlantic SST variability. However, the mechanisms of low-frequency variability of the AMOC and the links between AMOC variability and SST variability are poorly understood.

We propose to focus on two main science questions:

(1) On what timescales is the AMOC coherent with latitude and what does this coherence imply about mechanisms of AMOC variability and the predictability of the AMOC on various timescales?

(2) What is the relationship between AMOC variability and low-frequency SST variability? Does the AMOC play an active role in setting SST on interannual to decadal timescales and what are possible implications for predictability of Atlantic SSTs?

To answer these questions, we will conduct detailed analyses of dynamical mechanisms of AMOC variability in data-constrained, dynamically and kinematically consistent estimates of the global circulation (Estimating the Circulation and Climate of the Ocean, ECCO) and evaluate these mechanisms in multi-model comparisons (CMIP5 models). In the first year we will focus on an analysis of AMOC variability and its dynamical components in ECCO and a selection of CMIP5 models by (1) developing a three-dimensional partitioning of the flow field in terms of various dynamical components (e.g. the Gulf Stream, Gulf Stream recirculation region, near-surface Ekman transports, geostrophic interior, deep western boundary current, and eastern boundary current.) and (2) determining the meridional coherence of the AMOC/its dynamical components as a function of time and spatial scales. In the second year we will focus on the dynamical mechanisms of AMOC and SST/upper-ocean heat content (UOHC) variability in ECCO and CMIP5 models. We will determine modes of low-frequency AMOC variability and the associated patterns of SST/UOHC variability and use budget analyses to determine roles of air-sea heat fluxes, advection, mixing, etc in creating SST/UOHC anomalies. In the final year we will use ECCO to conduct directed model experiments to further explore mechanisms of AMOC and SST/UOHC variability. We will express mechanisms of AMOC variability hypothesized through our analysis of ECCO and CMIP5 models in terms of relationships between climate variables, which may be tested in other models. We will also determine how proposed mechanisms of AMOC variability may help explain the wide range of predictability of the AMOC/Atlantic SST seen in CMIP5 models.

This proposal is targeted at the NOAA Climate Program Office’s Earth System Science (ESS) competition: AMOC – Mechanisms and Decadal Predictability and is directly relevant to its objectives. Specifically, our proposal addresses three near-term research priorities highlighted in the recent SOST report “Assessing Meridional Overturning Circulation Variability: Implications for Rapid Climate Change”: (1) understanding the physical linkages between AMOC and Atlantic SST variability (2) combining the results of in-situ array programs and state estimates (3) conducting focused model inter-comparison studies in order to determine mechanisms of AMOC variability and their robustness. More broadly, our proposal is relevant to NOAA’s mission to help society cope with climate variability and change. The potential to predict climate on decadal timescales has significant economic and human health and safety applications. Our work furthers the new field of decadal prediction through (1) analysis of ocean state estimates that may prove useful for initializing decadal predictions and (2) study of the dynamics of important modes of climate variability, such as variability of the AMOC and Atlantic SST/UOHC.