Abstract's details
Performance of MERCATOR operational model at the Brazil Malvinas confluence
CoAuthors
Event: 2017 Ocean Surface Topography Science Team Meeting
Session: Science II: Large Scale Ocean Circulation Variability and Change
Presentation type: Type Poster
Contribution: not provided
Abstract:
Daily outputs from the operational Mercator global-ocean model (1/12) (PSY4) are analyzed in the complex region of the Brazil-Malvinas Confluence which is a stringent test for model outputs. The performance of the model is assessed by comparing velocity, temperature and salinity model outputs to in situ mooring time series gathered during the CASSIS-MALVINAS project. The data set includes 3 moorings (comprising an upward looking long-range ADCP deployed over the 1000 m isobaths) deployed at 41°S over the Patagonian Shelf break from December 2014 to December 2015.
In general, model means outputs and standard deviations are in good agreement with in situ data. For the along-slope velocity component, the agreement is particularly satisfying in the upper slope instruments where the correlation between along-slope velocity components of the model and 20-day low pass filter in situ velocities are larger than 0.7. During June 2015 the root mean square difference between the along-slope model velocities and the along-slope in situ velocities increases considerably (more than 20 cm/s) at the three mooring sites. The comparison of the sea level anomaly from altimetry data and from model outputs shows that the sea level anomaly is not correctly assimilated in the model during this period. The causes of this failure are discussed here.
The performance of the model is then assessed using the agreement with altimetry. We then use the model outputs to put the in-situ data in a larger spatial and temporal context (10 years), to derive quantities such as along-slope and across-slope transports, when accurate, and to investigate processes on the continental slope such as upwelling and downwelling events.
In general, model means outputs and standard deviations are in good agreement with in situ data. For the along-slope velocity component, the agreement is particularly satisfying in the upper slope instruments where the correlation between along-slope velocity components of the model and 20-day low pass filter in situ velocities are larger than 0.7. During June 2015 the root mean square difference between the along-slope model velocities and the along-slope in situ velocities increases considerably (more than 20 cm/s) at the three mooring sites. The comparison of the sea level anomaly from altimetry data and from model outputs shows that the sea level anomaly is not correctly assimilated in the model during this period. The causes of this failure are discussed here.
The performance of the model is then assessed using the agreement with altimetry. We then use the model outputs to put the in-situ data in a larger spatial and temporal context (10 years), to derive quantities such as along-slope and across-slope transports, when accurate, and to investigate processes on the continental slope such as upwelling and downwelling events.