Revisiting AMOC Transport Estimates From Observations and Models. Issue 10 (21st May 2021)

• Record Type: Journal Article
• Authors: Danabasoglu, Gokhan; Castruccio, Frederic S.; Small, R. Justin; Tomas, Robert; Frajka‐Williams, Eleanor; Lankhorst, Matthias
• Is Part Of: Geophysical research letters. Volume 48:Issue 10(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-05-21
• Subjects: Atlantic Meridional Overturning Circulation (AMOC) -- evaluation of transport methods in models | level‐of‐no‐motion and reference velocity -- MOVE and RAPID transport estimates; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1029/2021GL093045
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: Reference level assumptions used to calculate the Atlantic meridional overturning circulation transports at the RAPID and Meridional Overturning Variability Experiment (MOVE) observing arrays are revisited in an eddying ocean model. Observational transport calculation methods are complemented by several alternative approaches. At RAPID, the model transports from the observational method and the model truth (based on the actual model velocities) agree well in their mean and variability. There are substantial differences among the transport estimates obtained with various methods at the MOVE site. These differences result from relatively large and time‐varying reference velocities at depth in the model, not supporting a level‐of‐no‐motion. The methods that account for these reference velocities properly at MOVE produce transports that are in good agreement with the model truth. In contrast with the observational estimates, the model transport trends at MOVE and RAPID largely agree with each other on pentadal to multi‐decadal time scales. Plain Language Summary: Ocean transport arrays measure the mean and variability of ocean circulation. Due to technical, logistical, and financial constraints, these arrays do not measure ocean currents directly, but rather rely on a limited set of measurements leveraged by fundamental governing equations. Two observational arrays in the subtropical North Atlantic use these equations subject to some assumptions. In this study, a high‐resolution ocean model is employed to test these assumptions by estimating ocean transports in the model according to the methods applied at the two arrays. The model solutions do not support these assumptions at one of the sites. Nevertheless, there is an agreement in transport trends at both sites in the model, in contrast with the observational estimates. Proper determination of reference velocities in observations is important to advance understanding of basin‐wide circulation and its changes in the Atlantic Ocean. Key Points: Assumptions used in observational Meridional Overturning Variability Experiment (MOVE) and RAPID transport estimates are revisited in eddying model simulations Results reinforce the need for proper treatments of reference depths and velocities in geostrophic calculations in observations and models Simulated variability and trends at MOVE and RAPID locations are mostly consistent between the two sites, particularly on long time scales