QUALITY INFORMATION DOCUMENT For Global Ocean …cmems-resources.cls.fr/documents/QUID/CMEMS-GLO... · QUALITY INFORMATION DOCUMENT For Global Ocean Reanalysis Products GLOBAL-REANALYSIS-PHY-001-025

QUALITY INFORMATION DOCUMENT For Global Ocean Reanalysis Products GLOBAL-REANALYSIS-PHY-001-025

Issue: 3.5

Contributors: Gilles Garric, Laurent Parent

Approval Date by Quality Assurance Review Group :

Quality Information Document For products

GLOBAL-REANALYSIS-PHY-001-025

Ref :CMEMS-GLO-QUID-001-025 Date : Jan 2017 Issue : 3.5

© EU Copernicus Marine Service – Public Page 2/ 48

CHANGE RECORD

Issue Date § Description of Change Author Checked By

1.0 25 October 2011

all First version of document Nicolas Ferry, Laurent Parent, Simona Masina, Andrea Storto, Keith Haines, Maria Valdivieso, Bernard Barnier, Jean-Marc Molines

2.1 29 February 2012

V2.1 update Maria Valdivieso, Andrea Storto, Nicolas Ferry

2.2 14 January 2013

all Change product name Laurence Crosnier

2.3 11 February 2013

all Changes related to recommendations made by QuARG

Nicolas Ferry, Laurent Parent, Simona Masina, Andrea Storto, Keith Haines, Maria Valdivieso, Bernard Barnier, Jean-Marc Molines

2.4 27 September 2013

V2.4 update Nicolas Ferry, Laurent Parent, Simona Masina, Andrea Storto, Keith Haines, Maria Valdivieso, Bernard Barnier, Jean-Marc M li

2.5 19 May 2014

all New products and update for MyOcean V4.1

Hao Zuo, Andrea Storto

Yann Drillet

2.6 23 May 2014

header format correction Yann Drillet

2.7 7 July 2014 §1.2.8 Addition of EAN CGLORS

2.8 17 Dec 2014

all Addition of daily outputs Hao Zuo, Andrea Storto, Laurent Parent

2.9 March 2015 all Revision after V5 acceptance Yann Drillet

3.0 May 1 2015 all Change format to fit CMEMS graphical rules

L. Crosnier





3.1 January 2016

All Change format for CMEMS V2, suppress V1 products 001-004 and 001-010 no longer disseminated in V2, add year 2014 for GLORYS2V3 + a few cosmetic changes

M Drevillon Y Drillet

3.2 September 2016

CMEMS V2.2 UPDATE

Suppress product 001-009. Add product 001-025. Suppress MJM105b reference simulation and ARMOR3D reprocessed data in the validation framework of GLORYS2V4.

Gilles Garric, Laurent Parent

Y. Drillet

3.3 October Correction after review Marie Drevillon Y. Drillet

3.5 Jan 2017 Remove products 011 and 017 Yann Drillet Y. Drillet





TABLE OF CONTENTS

I EXECUTIVE SUMMARY ............................................................................................................................ 6 Products covered by this document ........................................................................................................... 6 I.1

Summary of the results ............................................................................................................................... 7 I.2 Temperature ....................................................................................................................................... 7 I.2.1 Salinity ................................................................................................................................................ 8 I.2.2 Ocean currents ................................................................................................................................... 8 I.2.3 Sea level............................................................................................................................................... 8 I.2.4 Transports .......................................................................................................................................... 9 I.2.5 Sea Ice concentration ......................................................................................................................... 9 I.2.6 Other ................................................................................................................................................... 9 I.2.7 Estimated Accuracy Number .......................................................................................................... 10 I.2.8

II Production Subsystem description .............................................................................................................. 11 GLORYS2V4 reanalysis (001_025) ........................................................................................................ 11 II.1

Summary ..................................................................................................................................................... 19 II.2

III Validation framework ............................................................................................................................. 20

IV Validation results of GLORYS2V4 ............................................................................................................. 21 Temperature ........................................................................................................................................... 21 IV.1

T-CLASS1-T3D-MEAN .................................................................................................................. 21 IV.1.1 T-CLASS1-SST-MEAN ................................................................................................................... 21 IV.1.2 T-CLASS1-SST-TREND ................................................................................................................. 22 IV.1.3 T-CLASS2-MOORINGS ................................................................................................................. 22 IV.1.4 T-CLASS3-SST_QNET_2DMEAN ................................................................................................ 24 IV.1.5 T-CLASS3-HC_LAYER ................................................................................................................. 24 IV.1.6 T-CLASS3-IC_CHANGE ............................................................................................................... 26 IV.1.7 T-CLASS4-LAYER ......................................................................................................................... 26 IV.1.8 T-DA-INNO-STAT .......................................................................................................................... 26 IV.1.9

Salinity ..................................................................................................................................................... 27 IV.2 S-CLASS1-S3D_MEAN .................................................................................................................. 27 IV.2.1 S-CLASS1-SSS-TREND .................................................................................................................. 28 IV.2.2 S-CLASS2-MOORINGS ................................................................................................................. 28 IV.2.3 S-CLASS3-SSS_SFW_2DMEAN ................................................................................................... 30 IV.2.4 S-CLASS3-SC_LAYER ................................................................................................................... 30 IV.2.5 S-CLASS3-IC_CHANGE ................................................................................................................ 31 IV.2.6 S-CLASS4-LAYER .......................................................................................................................... 32 IV.2.7 S-DA-INNO-STAT ........................................................................................................................... 32 IV.2.8

Currents .................................................................................................................................................. 33 IV.3 UV-CLASS1-15m_MEAN ............................................................................................................... 33 IV.3.1 UV-CLASS2-MOORINGS .............................................................................................................. 34 IV.3.2 UV-CLASS2-MKE_1000m ............................................................................................................. 36 IV.3.3

Sea level ................................................................................................................................................... 36 IV.4 SL-CLASS1-TREND_2D ................................................................................................................ 36 IV.4.1 SL-CLASS2-TIDE_GAUGE ........................................................................................................... 37 IV.4.2 SL-CLASS3-2DMEAN .................................................................................................................... 37 IV.4.3 SL-CLASS4 ...................................................................................................................................... 38 IV.4.4 SL-DA-INNO_STAT ....................................................................................................................... 38 IV.4.5

Transports ............................................................................................................................................... 39 IV.5 UV-CLASS3-VOL_TRANSP .......................................................................................................... 39 IV.5.1 UV-CLASS3-AMOC ....................................................................................................................... 40 IV.5.2 T-CLASS3-MHT .............................................................................................................................. 41 IV.5.3





Sea Ice ...................................................................................................................................................... 41 IV.6 SI-CLASS1-CONC_MEAN ............................................................................................................ 41 IV.6.1 SI-CLASS1-CONC_EXTREME .................................................................................................... 43 IV.6.2 SI-CLASS3-EXTENT ...................................................................................................................... 44 IV.6.3 SI-CLASS3-EXTENT_ANO ........................................................................................................... 45 IV.6.4 SI-CLASS3-VOL ............................................................................................................................. 45 IV.6.5 SI-CLASS3-VOL_ANO ................................................................................................................... 46 IV.6.6 SI-CONC-DA-INNO-STAT ............................................................................................................ 46 IV.6.7

Other ........................................................................................................................................................ 47 IV.7 OTH-CLASS1-MLD_CLIM_SEASON ......................................................................................... 47 IV.7.1





I EXECUTIVE SUMMARY

Products covered by this document I.1

The product assessed in this document is referenced as: GLOBAL-REANALYSIS-PHY-001-025 for Global Ocean Physics Reanalysis system product.

The goal of the CMEMS global ocean reanalysis is to provide a eddy permitting (1/4°) global ocean simulation, covering the recent period during which altimeter data are available (period starting with the launch of TOPEX POSEIDON and ERS-1 satellites early in the nineties), constrained by assimilation of observations and describing the space-time evolution of the following variables:

* sea_surface_height_above_geoid * sea_water_potential_temperature * sea_water_salinity * sea_water_x_velocity * sea_water_y_velocity * sea_ice_area_fraction * sea_ice_thickness * sea_ice_x_velocity * sea_ice_y_velocity

These reanalyses are built to be as close as possible to the observations (i.e. realistic) and in agreement with the model physics. It covers the “altimetric era” (namely 1st of January 1993 until 31 December 2015 for 001_025). The numerical products available for users are monthly mean averages describing the ocean from surface to bottom (5900 m).

The product GLOBAL-REANALYSIS-PHY-001-025 consists of a global ocean reanalyses datasets at ¼° horizontal resolution. The product is distributed on a regular ¼° grid. See the Product User Manual [3] for more details about the grid.

The table below (Table 1) synthesizes the different data sets of the global ocean physical reanalysis products:

Product Name Product origin

Production Unit

Data set Kind of data set

GLOBAL-REANALYSIS-

GLORYS2V4

GLO-MERCATOR-TOULOUSE-FR

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-monthly-t

reanalysis





PHY-001-025 dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-monthly-s

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-monthly-u-v

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-monthly-ssh

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-monthly-ice

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-coordinates

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-daily-t

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-daily-s

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-daily-u-v

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-daily-ssh

dataset-global-reanalysis-phy-001-025-ran-fr-glorys2v4-daily-ice

Table1: Overview of GLOBAL-REANALYSIS-PHYS-001_025 product.

Summary of the results I.2

Temperature I.2.1GLORYS2V4

GLORYS2V4 reanalysis has weak biases in temperature with respect to the Levitus 2009 climatology and in situ data (less than 0.4°C on average). Biases of GLORYS2v4 computed from these two datasets are of opposite signs. The largest biases occur in the [50m-100m] layer and in the northern Atlantic and Southern oceans.

The global mean sea surface temperature (SST hereafter) is close to the observations with a weak (warm) misfit of less than 0.5°C all along the reanalysis. The global positive SST linear trend is highly consistent with AVHRR data. The global mean surface net heat flux is weakly positive (+ 0.5 W.m2).

Along the equator, temperature profiles are very consistent with observations with RMSE (Root mean square errors) generally smaller than 0.4°C in the water column, apart from the [50m-200m] layers where the RMSE can intermittently reaches 1.2°C in the Indian Ocean. Correlations with equatorial moorings also decrease with depth.

The heat content constantly increases with time until 2011. Since 2011, the heat content may have stabilised. Only the deeper layers (below 2000m depth) show a constant cooling.





The thermal structure largely improves with the deployment of Argo buoys (after 2002), mainly in the upper layers (depth < 300m) with a (cold) bias close to 0°C and RMSE against all the in-situ observations less than 1°C. The RMSE also decreases with time in accordance with the observations network density.

Salinity I.2.2GLORYS2V4

The sea surface salinity (SSS hereafter) is generally fresher (less than 0.2 psu) than both the climatology and in situ data sets. However, largest fresh anomalies are found in tropical areas such as the western Atlantic and Pacific oceans. This surface fresh anomaly persists during the whole period covered by the reanalysis. Saltier surface waters are also found in the Arctic Ocean compared to the Levitus climatology. A general salty bias (less than 0.1 psu) is found in the [50m-200m] layer. Both biases and RMSE against all the in-situ observations drastically decrease with time in accordance with the increasing observations network density.

Positive trends of SSS are seen over most parts of the global ocean. Yet, local negative trends are found in the Arctic Ocean and in the Indonesian Throughflow. This overall positive trend is mitigated by a strong negative trend the first three years of the reanalysis.

Along the equator, salinity profiles are highly consistent with observations with RMSE generally smaller than 0.2 psu in the water column, apart at the surface where the RMSE is higher and can intermittently reach 0.3 psu °C in the Atlantic Ocean. Correlations with equatorial moorings also decrease with depth with however a generally weaker correlation at the surface compared to sub-surface scores.

The global salt content shows a weak positive trend with a somehow stabilisation during the last decade. Large interannual variability is found prior the full deployment of the Argo buoys.

Ocean currents I.2.3GLORYS2V4

GLORYS2V4 reanalysis reproduces well the main ocean currents. However, when compared to the AOML surface drifter climatology, the currents are generally underestimated, especially at mid and high latitudes. It is worth noting that this may be due to an artefact of the AOML surface current climatology as revealed by Grodsky et al. (2011).

Mean Kinetic Energy (MKE) at 1000m is in good agreement with the ANDRO Argo drift data base all along the boundary currents and the ACC.

The upper layers equatorial vertical dynamics structures are in well accordance with the moorings observations. The easterly surface current is however stronger in the western Pacific and the core of the Equatorial Under Current (EUC) is weaker. RMSE are generally smaller than 0.25 m.s-1 in the water column with a maximum in the lower part of the EUC. Correlations with equatorial moorings also generally decrease with depth.

Sea level I.2.4GLORYS2V4

GLORYS2V4 reanalysis is very close to altimetric observations and has a good ability to describe the sea level variability. Global and regional trends are very well reproduced. The globally averaged RMSE of the analysed sea level is about 6~7cm and the forecast sea level is 0.5cm higher. A global





and constant 0.5cm bias is present during the 2002-2006 time period. The GLORYS2V4 reanalysis and tide gauges are in relative agreement except along coastlines where tidal effects are important.

Transports I.2.5GLORYS2V4

Compared to Lumpkins and Speer (2007) estimates, GLORYS2V4 generally overestimates the mean volume transport in all basins and particularly all along the ACC current (by up to 30 Sv), probably due to a sharper N / S gradient of the CNES-CLS013 Mean Dynamic Topography (MDT) in that area. Four GLORYS2V4 transports in the northern Atlantic and Pacific oceans sections are in opposite sign compared to the Lumpkins and Speer (2007) estimates; however, they remain in the Lumpkins and Speer (2007)’s error bars. estimates.

The Atlantic Meridional Overturning Circulation (AMOC) shows no significant trend throughout the reanalysis period. It is 3 Sv weaker than the RAPID data, but exhibits high correlation (0.86) to these data.

The Meridional Heat Transport generally is in the error bars of the Ganachaud & Wunsch (2000) and Trenberth & Caron (2001) estimates.

Sea Ice concentration I.2.6GLORYS2V4

Thanks to sea ice concentration assimilation, all the sea ice extent variability (seasonal cycle, interannual variability and trend) is well reproduced in GLORYS2V4. Both sea ice extent and volume show a general decrease (resp. increase) in the Arctic Ocean (resp. around Antarctica). Biases and RMSE show a seasonal cycle, with a maximum during summertime, in both hemispheres. Biases are close to zero and RMSE is below 20% in both hemispheres, which is largely in the error bars of the observations.

Other I.2.7

GLORYS2V4

The mixed layer depth (MLD) of the GLORYS2V4 reanalysis exhibits similar patterns with those found in the De Boyer Montegut climatology data. However, GLORYS2V4 strongly overestimates the MLD in North Atlantic Ocean deep convection areas and lightly underestimates the MLD in the Southern Ocean during summertime.





Estimated Accuracy Number I.2.8

GLORYS2V4 1993-2015

Estimated Accuracy Numbers are calculated using mean and RMS errors from data assimilation statistics.

. Temperature (Celsius degrees)

Level Period 1993-2015 1993-2005 2005-2015 Mean RMS Mean RMS Mean RMS 0-100 -0.03 0.78 -0.04 0.87 -0.01 0.68 100-300 -0.02 0.69 -0.04 0.77 -0.01 0.61 300-800 -0.01 0.43 -0.01 0.49 -0.01 0.36 800-2000 0.01 0.18 0.01 0.24 0.00 0.12

Salinity (psu)

Level Period 1993-2015 1993-2005 2005-2015 Mean RMS Mean RMS Mean RMS 0-100 0.002 0.212 0.003 0.254 0.001 0.165 100-300 0.007 0.110 0.009 0.126 0.005 0.092 300-800 0.001 0.059 0.003 0.068 0.000 0.049 800-2000 0.001 0.031 0.002 0.040 0.000 0.020

Sea Level (cm)

Period 1993-2015 1993-2005 2005-2015 Mean RMS Mean RMS Mean RMS 0.02 5.7 0.07 5.6 -0.03 5.8

SST°C

Period 1993-2015 1993-2005 2005-2015 Mean RMS Mean RMS Mean RMS -0.07 0.6 -0.1 0.6 -0.05 0.5





II PRODUCTION SUBSYSTEM DESCRIPTION

Production Center: GLOBAL MFC

Production Unit:

- GLORYS2V4: Mercator Ocean

GLORYS2V4 reanalysis (001_025) II.1

GLORYS2V4 reanalysis relies on three main components:

1. the ocean model (including the surface atmospheric boundary condition)

2. the data assimilation method

3. the assimilated observations

We review here these three components.

Ocean model:

- OGCM configuration:

The reanalysis system is based on the version 3.1 of NEMO (Nucleus for European Models of the Ocean) ocean model (Madec, 2008); the configuration uses the tripolar ORCA grid type (1442x1021 grid points) (Madec and Imbard, 1996) at ¼° horizontal resolution (the three poles being located in Antarctica, Northern Asia and Northern Canada). The ¼ ° resolution of the ORCA grid type (ORCA025 hereafter) corresponds to 27 km at the equator, 22 km at Cape Hatteras (Gulf Stream area) and 12 km at high latitudes (Arctic Ocean & regional seas around Antarctica).

The vertical grid has 75 z-levels, with a resolution of 1 meter near the surface, 200 meters in the deepest ocean (> 3000m depth) and 24 levels with the upper 100m.

The bathymetry used in the system is a combination of interpolated ETOPO1 (Amante and Eakins, 2009) and GEBCO8 (Becker et al., 2009) databases. ETOPO datasets are used in regions deeper than 300 m and GEBCO8 is used in regions shallower than 200 m with a linear interpolation in the 200 m – 300 m layer. The minimum depth in the model is set to 12 meters, except in the Bahamas region (~ 3 meters). Three islands have been added in the Torres Strait to diminish the transport and the Palk Strait (between India and Sri Lanka) has been opened with a 3m depth.

A “partial cells” parameterization (Adcroft et al., 1997) is chosen to represent the topographic floor as staircases but making the depth of the bottom cell variable and adjustable to the real depth of the ocean floor (Barnier et al., 2006).

The horizontal pressure gradient is represented by a free surface formulation in which high frequency gravity waves are filtered out (Roullet and Madec, 2000).

The advection of the tracers (temperature and salinity) is computed with a total variance diminishing (TVD) advection scheme (Lévy et al., 2001; Cravatte et al., 2007). A Laplacian lateral isopycnal diffusion on tracers is used (300m2.s-1 at the Equator and decreasing poleward proportionally to the grid size).

The momentum advection term is computed with the energy and enstrophy conserving scheme proposed by Arakawa and Lamb (1981).





An horizontal bi-harmonic viscosity for the lateral dissipation of momentum (−1×1011 m4 s−1 at the Equator and decreasing poleward as the cube of the grid size) is used. In this system, a Laplacian operator (2000 m2.s-1) is applied in the Ob and Ienissei estuaries to avoid local numerical instabilities.

The vertical mixing is parameterized according to a turbulent closure model (order 1.5) adapted by Blanke and Delecluse (1993). Internal-tide driven mixing is parameterized following Koch-Larrouy et al. (2007) for tidal mixing in the Indonesian Seas.

The lateral friction condition is a partial-slip condition with a regionalisation of a no-slip condition (over the Mediterranean Sea).

Instead of being constant, the depth of light extinction is separated in Red-Green-Blue bands depending on the chlorophyll data distribution from mean monthly SeaWIFS climatology.

The reanalysis system uses the thermodynamic-dynamic sea-ice model LIM2 (Fichefet and Maqueda, 1997) with the Elastic-Viscous-Plastic rheology (Hunke and Dukowicz, 1997). The thermodynamics are based on the 2+1 (ice+snow) layers formulation.

Surface forcing fields

The reanalysis system is driven at the surface by the ERA-Interim reanalysis products (Dee et al., 2011), interpolated onto the ORCA025 grid using the Akima interpolation algorithm. Momentum and heat turbulent surface fluxes are computed from the Large and Yeager (2004) CORE bulk formulae using the following set of atmospheric variables at 3H sampling: surface air temperature and surface humidity at 2 m height, mean sea level pressure and wind at 10 m height. Daily downward longwave and shortwave radiative fluxes and rainfall (solid + liquid) fluxes are also used in the surface heat and freshwater budgets. An analytical formulation (Bernie et al., 2005) is applied to the daily shortwave flux in order to reproduce an ideal diurnal cycle. This parameterization aims at better representing the night-time convection which takes place in the upper most layer of the ocean (Bernie et al., 2007).

Due to large known biases in precipitations and radiative fluxes at the surface, a satellite-based large-scale correction (Garric et al., 2011) is applied to the ERA-Interim fluxes. This correction is applied (1) to the low-passed filtered ERA-Interim fluxes (80% amplitude attenuation at the synoptic scales) in order to avoid any changes on synoptic patterns such as cyclones and (2) to the seasonal climatology in order to avoid any changes of the interannual signal. No corrections are applied at high latitudes (poleward 65°N and 60°S).

Also, specific corrections are applied at high latitudes to counteract knownn warm biases in ERA-Interim surface conditions (Jakobson et al. 2012, Lupkes et al. 2010). In the Arctic Ocean northward of 80°N: surface air temperature is cooled by 2°C and 85% of air humidity is retained. Southward of 60°S in the Southern Ocean: 80% of downward ERAinterim short-wave radiation is retained and 110% of downward ERAinterim long-wave radiation is applied.

Furthermore, in order to avoid artificial drifts of the globally-averaged sea-surface height due to the unbalanced fresh water budget, the evaporation minus precipitation quantities (EmP hereafter) is set equal to zero at each model time-step. In addition, a global mean EmP seasonal cycle is imposed to the one observed by the GRACE geodetic mission.

No global restoring strategy has been implemented in this system to sea surface salinity or to the sea surface temperature. However, a temperature and salinity 3D-restoring towards EN4 products (Good et al., 2013) is applied below 2000m and poleward of 60°S with a representative time scale of 20 years. This latter restoring has been applied to counteract the possible drift of the volume transport in the Antarctic Circumpolar Current.

Initial Conditions:

The simulation started at rest the 4th December 1991, with prescribed temperature and salinity based on EN4 (Good et al., 2013).





Initial conditions for sea ice (ice concentration) were inferred from the NSDIC Bootstrap products for December 1991. The ice thickness has been issued from analytical relationships found between a mean (2002-2009) GLORYS1V1 January sea ice concentration and its ice thickness counterpart. This method allows to considerably reducing the sea ice mass spin up.

Time step and ramp up phase: the time step is set to 720 sec the first and second weeks of the run; then, the time step is set to 1080s all along the experiment. No data are assimilated during the first two weeks.

Data Assimilation method

The data are assimilated by means of a reduced-order Kalman filter, based on the SEEK formulation introduced by Pham et al. (1998), with a 3-D multivariate modal decomposition of the forecast error and a 7-day assimilation cycle. Here we present a short description of what we call Système d’Assimilation Mercator version 2 (SAM2).

An important feature of the GLORYS2V4 reanalysis system (implemented also in the real-time global forecasting systems) is that the analysis is performed at the middle of the 7-day assimilation cycle, and not at the end of the assimilation cycle. Doing so, future and past observations are used in the analysis, providing a better estimate of the ocean because the analysis is temporally centred, with respect to the observations used.

The SEEK formulation requires knowledge of the forecast error covariance of the control vector. Altimeter data, in situ temperature and salinity vertical profiles and satellite sea surface temperature are jointly assimilated to estimate the initial conditions for numerical ocean forecasting. Satellite sea ice concentration is assimilated in the system in a monovariate/monodata mode. The forecast error covariance is based on the statistics of a collection of 3D ocean state anomalies (typically a few hundred) and is seasonally variable (i.e. fixed basis, seasonally variable). This approach is similar to the Ensemble optimal interpolation (EnOI) developed by Oke et al., (2008) which is an approximation to the EnKF that uses a stationary ensemble to define background error covariances. In our case, the anomalies are high pass filtered ocean states (Hanning filter, length cut-off frequency = 1/30 days-1) available over the 1993-2009 time period every 3 days. These ocean states come from a reference simulation in which no data assimilation is performed.

A 3D-VAR bias correction method has been implemented to correct large-scale temperature and salinity biases when enough observations are present. The bias correction corrects the slowly varying large scale error of the model whereas SAM2 assimilation scheme corrects the smaller scales of the model forecast error.

These increments are applied with an Incremental Analysis Update (Bloom et al., 1996, Benkiran and Greiner, 2008). The IAU is a low-pass filter, which gives a smooth model integration, without jumps induced at the analysis time when the increment is applied on one time step only (classical model correction). The IAU also reduces spin up effects after the analysis time. In practice, the IAU scheme is more costly than the “classical” model correction because of an additional model integration over the assimilation window.

Assimilated observations:

The assimilated data consist of satellite SST and SLA data, in situ temperature and salinity profiles, and sea ice concentration. The model innovation (observation minus model equivalent, y-Hx) is computed at appropriate time (FGAT, First Guess at Appropriate Time).

As stated in the model description section, the start time for GLORYS2V4 reanalysis is the 4th December 1991 at 00H UTC. The ocean model is integrated during the first two weeks without data assimilation. Then, all available observations listed below are assimilated.

SLA: provided by CMEMS SLA TAC (CLS: Collecte Localisation Satellites): SEALEVEL-GLO-SLA-L3-REP-OBSERVATIONS-008-001-b product

Here is a short summary of the Delayed Time altimetric data sets availability:





Jason-2 Jason-1 new (1)

Jason-1 GFO Envisat Envisat new (2)

ERS-1(3) ERS-2

Temporal Time availability: beginning / end

2008/10

2009/02

2012/03

2002/04

2008/10

2000/01

2008/09

2002/10

2010/10

2010/10

2012/04

1992/10

2002/10

Topex new (4)

Topex Cryosat2

AltiKa H2-YA2

Temporal Time availability: beginning / end

2002/09

2005/10

1992/09

2002/04

2011/01 2013/03 2014/04

2016/01

Table: Time period during which altimetric data sets are available

(1) Jason-1 new orbit: starting 2009/02

(2) Envisat new orbit: starting 2010/10

(3) There are no ERS-1 data between December 23, 1993 and April 10, 1994 (ERS-1 phase D – 2nd ice phase)

(4) T/P new orbit: starting 2002/09

The assimilation of SLA observations requires the knowledge of a mean sea surface height. This surface reference, closely linked to the mean circulation / thermodynamical heat and salt content of the ocean, is an adjusted version of Mean Dynamic Topography (MDT hereafter) dataset. SST: The daily Reynolds 1/4° AVHRR-only daily SST version 2 product (see http://www.ncdc.noaa.gov/oa/climate/research/sst/oi-daily.php) is assimilated once at the date of the analysis (i.e. the day 4 at 24H00 of the assimilation cycle).

In Situ profiles (T, S): The CORA data base provided by CMEMS in situ TAC (Coriolis data centre) is assimilated (http://www.coriolis.eu.org/Science/Data-and-Products/CORA-03) in GLORYS2V4.

Sea Ice concentration: The IFREMER/CERSAT products (Ezraty et al., 2007) are assimilated once at the date of the analysis (i.e. the day 4 at 24H00 of the assimilation cycle).

Observation error:

Observation error includes both the instrumental error (Rme) and the model representativeness error (Rre). These errors are supposed to be uncorrelated and are modelled by diagonal matrices (i.e. the observation errors are mutually uncorrelated). SST: An empirical error is used. It is function of the representativeness error Rre of in-situ measurements and equal to RSST = max(0,5°C, Rin_situ(T)).

SLA: The SLA observation error Rme (in variance) is specified according to the knowledge of instrumental features. We use a 2 cm error for JASON (with new orbit or not) and TOPEX-POSEIDON (with new orbit or not) and a 3.5 cm error for ERS-2, GFO and ENVISAT. In October 2010, the Envisat altimeter was brought to a lower orbit, which has led to a slight degradation of data quality (Ollivier and Faugere, 2010). In consequence, a 5.5 cm error is attributed to this altimeter (Envisat new). We use a 4 cm error for Cryosat-2, 2 cm error for Altika and 4 cm error for HY-2A.

http://www.ncdc.noaa.gov/oa/climate/research/sst/oi-daily.php

http://www.coriolis.eu.org/Science/Data-and-Products/CORA-03





Mean Dynamic Topography: A representativeness error for the MDT has been added to the SLA observation error. This error is 5 cm on average. Largest values (~10 cm) are located on shelves, along the coast (model representativeness + observation error due to tides) and in the regions of sharp fronts (Gulf Stream, Kuroshio,...).

In situ profiles: The observation error of temperature and salinity profiles includes both the instrument error and the representativeness error by assuming that the model representativeness is a function of the geographical location and depth. For temperature, this error is dominated by the inaccuracy of the thermocline position given by the model and the data (profilers depth uncertainty can reach 5 m or 2 % of the depth). It ranges from about ~1°C at the surface to 0.1°C below 500m. The higher surface values correspond to mixed layer processes or thermocline motions that are not resolved by the model (filaments, internal gravity waves …). Contrary to the temperature, the salinity observation error can be very large in coastal areas due to run-off uncertainty. The halocline gradient is also weaker than the thermocline gradient, leading to a more homogeneous error on the vertical.

Sea Ice concentration: An empirical error is used. The error is 0.01 when the concentration is equal to 0, then a linear decreasing trend is applied from 0.25 (concentration equal to 0.01) to 0.05 (concentration equal to 1).

Changes compared to the previous GLORYS (GLORYS2V3) system:

• New initial conditions for temperature and salinity based on EN4 (Good et al., 2013) are replacing the Levitus (98) climatology used in the previous system. A method using a robust regression model applied to the EN4 monthly objective analysis allowed to re-built the December 1991 water masses conditions. This method considerably reduced the imbalance between a climatology not centred on the initial date and the steric signal seen by the altimetry in 1992.

• The monthly seasonal climatology river runoff is now inferred from Dai et al. (2009). It is a reliable estimate for the world’s 925 largest rivers of continental freshwater discharge. This database uses new data, mostly from recent years, streamflow simulated by Community Land Model version 3 to fill the gaps, in all lands areas except Antarctica and Greenland. In addition, the runoff fluxes coming from Greenland and Antarctica ice sheets and glaciers melting has been built from the Altiberg icebergs database project (Tournadre et al., 2012). This complements the estimation of Silva et al. (2006) for Antarctica that was already present in the previous system. The total river runoff represents 1.3 Sv.

• Large-scale corrections of precipitations, resp. radiatives fluxes, at the surface are now performed with the Passive Microwave Water Cycle (PMWC) (Hilburn, 2009), resp GEWEX SRB 3.1 (http: //eosweb.larc.nasa.gov/PRODOCS/srb/table srb.html), satellite data sets.

• A temperature and salinity restoring toward EN4 products (Good et al., 2013) has been added at Gibraltar and Bab-el-Mandeb straits to correctly capture the outflow of Mediterranean and Red seas waters into the Atlantic and Indian oceans.

• The temperature and salinity 3D-restoring applied below 2000m and poleward of 60°S with a representative time scale of 20 years is now made towards the EN4 products.

• Despite the previous corrections and updates, the freshwater budget (EmP) is far from being balanced. In order to avoid any mean sea-surface-height drift due to the poor water budget closure, we add a trend of 1.4 mm/year to the runoffs in order to represent somehow the recent estimation of the global water mass addition to the ocean (from glaciers, land water storage, Greenland & Antarctica ice sheets) (Chambers et al., 2016).

• The sparsity of the observation networks (both altimetry and in situ) during the 7-days assimilation window together with the uncertainties in the MDT estimation are not able to correctly estimate the mean global sea level; we then set to zero the global mean increment of the steric sea surface height.

• A new MDT based on the “CNES-CLS13” MDT with adjustment using high resolution analysis and with improvement on the Post Glacial Rebound (or Glacial Isostatic Adjustment, e.g. Peltier 2004), has been used. This new product also takes into account the last version of the GOCE geoid and is replacing the MDT named “CNES-CLS09” derived from observations (Rio et al., 2011) that was used in the previous system.

• A new 20-year altimeter reference period has been used in place of the previous 7-year reference period for assimilated SLA products.





• The New CORA 4.1 in situ database produced by the CORIOLIS centre (Szekely et al., 2015) is now assimilated (http://www.coriolis.eu.org/Science/Data-and-Products). Compared to the CORAv3.3 release, the CORA4.1 database includes the temperature and salinity vertical profiles from the sea mammals database (33 000 profiles from elephant seals) (Roquet et al., 2011), as well as moorings from TAO/RAMA/PIRATA programs and corrections on XBT measurements. This data set has been extensively quality controlled using classical “in situ” quality control procedures. For years 2014 and 2015, the delayed mode database was not available, so that the real-time database produced by the Coriolis data centre was used instead.

References :

Adcroft, A., C. Hill, and J. Marshall, (1997) Representation of topography by shaved cells in a height coordinate ocean model, Mon. Weather Rev., 125, 2293-2315.

Amante, C. and B.W. Eakins (2009), ETOPO1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis, NOAA Technical Memorandum NESDIS NGDC-24, 19 pp.

Arakawa, A. and Lamb, V. R. (1981): A potential enstrophy and energy conserving scheme for the shallow water equations, Mon. Weather. Rev., 109, 18–36.

Barnier, B., Madec, G., Penduff, T., Molines, J. M., Treguier, A.M., Le Sommer, J., Beckmann, A., Biastoch, A., B¨oning, C., Dengg, J., Derval, C., Durand, E., Gulev, S., Remy, E., Talandier, C., Theetten, S., Maltrud, M., McClean, J., and De Cuevas, B.(2006), Impact of partial steps and momentum advection schemes in a global circulation model at eddy permitting resolution, Ocean Dynam., 56, 543–567.

Becker JJ, Sandwell DT, Smith WHF, Braud J, Binder B, Depner J, Fabre D, Factor J, Ingalls S, Kim SH, Ladner R, Marks K, Nelson S, Pharaoh A, Trimmer R, Von Rosenberg J, Wallace G, Weatherall P (2009) Global bathymetry and elevation data at 30 arc seconds resolution: SRTM30_PLUS. Mar Geod 32:355–371. doi:10.1080/01490410903297766.

Bernie D.J., Guilyardi E., Madec G., Slingo J.M., Woolnough S.J. (2007), Impact of resolving the diurnal cycle in an ocean-atmosphere GCM. Part 1: a diurnally forced OGCM. Clim. Dyn.29:575-590, doi:10.1007/s00382-007-0249-6.

Bernie, D. J., Woolnough, S. J., Slingo, J. M., and Guilyardi, E. (2005), Modeling diurnal and intraseasonal variability of the ocean mixed layer, J. Climate, 18, 1190–1202.

Benkiran M.,Greiner E. (2008), Impact of the Incremental Analysis Updates on a Real-Time System of the North Atlantic Ocean. Journal of Atmospheric and Oceanic Technology 25(11): 2055.

Blanke, B. and Delecluse, P. (1993), Variability of the tropical Atlantic-Ocean simulated by a general-circulation model with 2 different mixed-layer physics, J. Phys. Oceanogr., 23, 1363–1388.

Bloom, S.C., Takas, L.L., Da Silva, A.M., Ledvina, D. (1996), Data assimilation using incremental analysis updates. Monthly Weather Review, 124, 1256–1271.

Chambers, D.P., A. Cazenave, N. Champollion, H. Dieng, W. Llovel, R. Forsberg, K. von Schuckmann, and Y. Wada, 2016: Evaluation of the global mean sea level budget between 1993 and 2014. Surv. Geophys., early on-line, doi:10.1007/s10712-016-9381-3.

Cravatte, S., Madec, G., Izumo, T., Menkes, C., and Bozec, A. (2007), Progress in the 3-D circulation of the eastern equatorial Pacific in a climate, Ocean Model., 17, 28–48.

Dai A., T. Qian, K. Trenberth and J.D. Milliman (2009), Changes in Continental Freshwater Discharge from 1948 to 2004, J. Climate, vol. 22, p.2773-2792.D.P. Dee and S.M. Uppala and A.J. Simmons and P. Berrisford and P. Poli and S. Kobayashi and U. Andrae and M.A. Balmaseda and G. Balsamo and P. Bauer and P. Bechtold and A.C.M. Beljaars and L. van de Berg and J. Bidlot and N. Bormann and C. Delsol and R. Dragani and M. Fuentes and A.J. Geer and L. Haimberger and S.B. Healy and H. Hersbach and E.V. Holm and L. Isaksen and P. Kallberg and M. Kohler and M. Matricardi and A.P. McNally and B.M. Monge-Sanz and J.-J. Morcrette and B.-K. Park and C. Peubey and P. de Rosnay and C. Tavolato and J.-N. Thepaut and F. Vitart (2011), The ERA-Interim





reanalysis: configuration and performance of the data assimilation system, Quart. J. Roy. Met. Soc., vol. 137, doi: 10.1002/qj.828, p.553-597.

Ezraty, R., Girard-Ardhuin, F., Piolle, J. F., Kaleschke, L., and Heygster, G.: Arctic and Antarctic sea ice concentration and Arctic sea ice drift estimated from Special Sensor Microwave data, User’s Manuel Version 2.1, CERSAT, 2007.

Fichefet, T., and M.A. Morales Maqueda, 1997 : Sensitivity of a global sea ice model to the treatment of ice thermodynamics and dynamics. Journal of Geophysical Research, 102, 12,609-12,646.

Garric G., Verbrugge N, and C Bricaud (2011), Large scale ERAinterim radiative and precipitation surface fluxes assessment, correction and application on 1/4° global ocean 1989-2009 hindcasts. EGU2011-10892. XY136, EGU 2011.

Good, S. A., M. J. Martin and N. A. Rayner, (2013) EN4: quality controlled ocean temperature and salinity profiles and monthly objective analyses with uncertainty estimates, J. of Geoph. Research: Oceans, 118, 6704-6716, doi:10.1002/2013JC009067

Grodsky, S.A., R. Lumpkin and J.A. Carton, 2011: Spurious trends in global surface drifter currents. GEOPHYSICAL RESEARCH LETTERS, VOL. 38, L10606, doi:10.1029/2011GL047393, 2011

Hilburn, K. A., (2009), The passive microwave water cycle product, report number 072409, Remote Sensing Systems, 30 pp.

Hunke, E.C., Dukowicz, J.K., (1997), An elastic–viscous–plastic model for sea ice dynamics. Journal of Physical Oceanography 27, 1849–1867

Jakobson, E., Vihma, T., Palo, T., Jakobson, L., Keernik, H. and Jaagus, J. 2012, Validation of atmospheric reanalyses over the central Arctic Ocean, Geophys. Res. Lett., 39(L10802), doi:10.1029/2012GL051591.

Koch-Larrouy, A., G. Madec, P. Bouruet-Aubertot, T. Gerkema, L. Bessieres, and R. Molcard, 2007 : Tidal mixing in the indonesian seas and its effect on the tropical climate system. Geophys. Res. Let., 34, L04 604, doi :10.1029/2006GL028405, URL http://dx.doi.org/10.1029/2006GL028405.

Large,W. G. and S. Yeager ( 2004), Diurnal to decadal global forcing for ocean and sea-ice models : the data sets and flux climatologies. NCAR Technical Note, NCAR/TN-460+STR, CGD Division of the National Center for Atmospheric Research.

Lévy, M., Estublier, A., and Madec, G. (2001), Choice of an advection scheme for biogeochemical models, Geophys. Res. Lett., 28, 3725–3728, doi:10.1029/2001GL012947.

Lüpkes, C., Vihma, T., Jakobson, E., Konig-Langlo, G. and Tetzlaff, A. 2010: Meteorological observations from ship cruises during summer to the central Arctic: A comparison with reanalysis data, Geophys. Res. Lett., 37(L09810), doi:10.1029/2010GL042724.

Madec, G., (2008), NEMO ocean engine. Note du Pôle de modélisation, Institut Pierre-Simon Laplace(IPSL), France, No 27, ISSN No 1288-1619.

Madec, G. and M. Imbard (1996), A global ocean mesh to overcome the North Pole singularity, Climate Dynamics, vol.12, p.381-388.

Ollivier, A. and Y Faugere (2010), Envisat RA-2/MWR ocean data validation and cross-calibration Activities, Yearly report, Technical Note CLS.DOS/NT/10.018, Contract N SALP-RP-MA EA-21920-CLS.

Peltier W.R (2004) Global Glacial Isostasy and the Surface of the Ice-Age Earth: The ICE-5G(VM2) model and GRACE. Ann. Rev. Earth. Planet. Sci.. 32,111-149.

Pham, D., J. Verron, and M. Roubaud (1998), A singular evolutive extended Kalman filter for data assimilation in oceanography. J. Mar. Syst., 16, 323–340.

Oke, P. R., G. B. Brassington, D. A. Griffin, and A. Schiller (2008), The Bluelink Ocean Data Assimilation System (BODAS), Ocean Modell., 21, 46– 70, doi:10.1016/j.ocemod.2007.11.002.





Rio, M. H., S. Guinehut, and G. Larnicol (2011), New CNES-CLS09 global mean dynamic topography computed from the combination of GRACE data, altimetry, and in situ measurements, J. Geophys. Res., 116, C07018, doi:10.1029/2010JC006505.

Roullet, G. and G. Madec, 2000 : salt conservation, free surface, and varying levels : a new formulation for ocean general circulation models. J. Geophys. Res, 105, 23,927–23,942.

Roquet F., J-B Charrassin, S. Marchand, L. Boehme, M. Fedak, G. Reverdin and C. Guinet (2011), Delayed-mode calibration of hydrographic data obtained from animal-borne satellite relay data loggers, J. Atmos. Ocean. Tech., 28, 787-801.

Silva, T. A. M., Bigg, G. R., and Nicholls, K. W. (2006), Contribution of giant icebergs to the Southern Ocean freshwater flux, J. Geophys. Res., 111, C03004, doi:10.1029/2004JC002843.

St. Laurent, L., H. Simmons, and S. Jayne, 2002 : Estimating tidally driven mixing in the deep ocean. Geophys. Res. Let., 29, 2106, doi :10.1029/2002GL015633,URL http://dx.doi.org/10.1029/2002GL015633.

Szekely, T., Gourrion, J., Brion, E., Von Schuckmann, K. Reverdin, G., Grouazel, A., and Pouliquel, S., 2015: CORA4.1: A delayed-time validated temperature and salinity profiles and timeseries product. Proceeding from 7e EuroGOOS conference, To be published.

Tournadre J., Girard-Ardhuin F. and B. Legrézy (2012), Antarctic icebergs distributions, 2002–2010, J. Geo. Res.-Oceans, 117-C5, doi: 10.1029/2011JC00744.





Summary II.2

System name

domain resolution

Model version Data Assimilation

Assimilated Observations

GLORYS2V4

Global, ¼°, 75 vertical levels

ORCA025 LIM2 EVP NEMO 3.0, 3-hourly atmospheric forcing from ERA-Interim, including SW+LW+PRECIP corrections Date of start: 4th December 1991. Model spin-up: 2 weeks

SAM2 (SEEK Kernel) + FGAT + IAU and bias correction

Reynolds 0.25° AVHRR-only-SST, DT SLA from all altimetric satellites, in situ T, S profiles from CORIOLIS CORAv3.3 data base, CERSAT Sea Ice Concentration





III VALIDATION FRAMEWORK

The validation methodology defined and used to validate the global ocean reanalysis products is described on documents [1] and [2] and consists in a series of diagnostics defined in common by all global ocean reanalysis / reference forced simulation producers.

The reference validation period is 1st January 1993 – 31 December 2011 (the period initially spanned by the reanalysis). The times series are based on monthly fields except for innovation diagnostics (see T-DA_INNO_STAT, S-DA_INNO_STAT, SL-DA-INNO_STAT, SI-CONC-DA-INNO_STAT sections). The reanalysis product has been extended until 2015 (001_025). Some of the metrics (mostly average error time series such as innovation diagnostics) have been updated so that the user can check that the quality stays of the same order of magnitude over those last years.





IV VALIDATION RESULTS OF GLORYS2V4

Temperature IV.1

T-CLASS1-T3D-MEAN IV.1.1

Fig. 1.1: Differences of temperature (in °C) between GLORYS2V4, and the World Ocean Atlas 2009 climatology (WOA09) at 0.5m, 97m, 300m, 773m and 1945m depth, respectively. For this comparison, model data have been averaged over the period 1993 – 2015.

T-CLASS1-SST-MEAN IV.1.2





Fig. 1.2: Mean 1993-2015 difference in sea surface temperature (°C) between GLORYS2V4 and the AVHRR SST data.

T-CLASS1-SST-TREND IV.1.3

Fig. 1.3: Linear 1993-2015 SST trend of AVHRR data (left panel) and GLORYS2V4 (right panel).

T-CLASS2-MOORINGS IV.1.4





TAO/TRITON 0N-147E TAO/TRITON 0N-165E TAO/TRITON 0N-140W TAO/TRITON 0N-110W

PIRATA 0N-23W PIRATA 0N-0E RAMA: 0N-90E





Fig. 1.4: Mean equatorial temperature profile (top), RMSE (middle), correlation (bottom) estimated over the period 1993 – 2015: observations (black) and GLORYS2V4 (green). Moorings observations are coming from TAO/TRITON: 0N-140W, 0N-110W,0N-165E,0N-147E, PIRATA:0N-23W,0N-0E and RAMA: 0N-90E.

T-CLASS3-SST_QNET_2DMEAN IV.1.5

Fig. 1.5: Global spatial average of SST (left, °C) and surface net heat flux (right, W.m-2) over the period 1993 – 2015: observations (black), GLORYS2V4 (green).

The long term mean value of the surface net heat flux in GLORYS2V4 is +0.5 W.m-2,

T-CLASS3-HC_LAYER IV.1.6





Fig. 1.6: Time evolution of global average of temperature in different layers [0-800m], [0-2000m], [2000m-bottom] and [0-bottom] layers for the Levitus98 climatology repeated for the entire period (black) and GLORYS2V4 (green).





T-CLASS3-IC_CHANGE IV.1.7

Fig. 1.5: Time evolution of spatially averaged GLORYS2V4 temperature change with respect to initial conditions over the period 1993 – 2015.

T-CLASS4-LAYER IV.1.8Not available yet for GLORYS2V4

T-DA-INNO-STAT IV.1.9





Fig. 1.9: Time evolution of the domain averaged innovation in terms of bias (left) and RMSEr (right): SST (top) and CORA 4.1 in-situ temperature profiles (bottom). Top: AVHRR SST data (black), CORA4.1 SST data (blue) (in °C). Covered time period: 1993 – 2012 and 1993-2015.

Salinity IV.2

S-CLASS1-S3D_MEAN IV.2.1





Fig. 2.1: Differences of salinity (in psu) between GLORYS2V4 and the World Ocean Atlas 2009 climatology (WOA09) at 0.5m, 97m, 300m, 773m and 1945m depth. GLORYS2V4 data have been averaged over the period 1993 – 2015.

S-CLASS1-SSS-TREND IV.2.2

Fig. 2.2: Linear 1993-2015 SSS trend of GLORYS2V4.

S-CLASS2-MOORINGS IV.2.3











Fig. 2.3: Mean equatorial salinity profile (top), RMSE (middle) and correlation (bottom) estimated over the period 1993 – 2015: observations (TAO/TRITON: 0N-140W, 0N-110W,0N-165E,0N-147E, PIRATA:0N-23W,0N-0E and RAMA: 0N-90E) (black) and GLORYS2V4 (green).

S-CLASS3-SSS_SFW_2DMEAN IV.2.4

Fig. 2.4: Global spatial average of SSS from (left, psu) and surface net fresh water heat flux (right, mm.day-1) over the period 1993 – 2015 from GLORYS2V4 (green). The long term mean value of the net surface fresh water in GLORYS2V4 is -0.005 mm/day-1,

S-CLASS3-SC_LAYER IV.2.5





Fig. 2.5: Time evolution of global spatial average of salinity (psu) in [0-800m], [0-2000m], [2000m-bottom] and [0-bottom] layers for the Levitus98 climatology repeated for the entire period (black) and GLORYS2V4 (green).

S-CLASS3-IC_CHANGE IV.2.6GLORYS2V4





Fig. 2.6: Time evolution of GLORYS2V4 salinity changes with respect to initial conditions over the period 1993 – 2015.

S-CLASS4-LAYER IV.2.7

Not available yet for GLORYS2V4.

S-DA-INNO-STAT IV.2.8

Fig. 2.8: Time evolution of the domain average innovation in terms of bias (left) and RMSE(right) for CORA 4.1 in-situ salinity profiles. Covered time period: 1993 - 2015.





Currents IV.3

UV-CLASS1-15m_MEAN IV.3.1





Fig. 3.1: Zonal (top) and meridional (bottom) 15m depth current climatology. NOAA AOML drifter-derived climatology (left) and GLORYS2V4 (right). GLORYS2V4 model data have been averaged over the period 1993 – 2015.

UV-CLASS2-MOORINGS IV.3.2







Data not available

Data not available





Data not available

Fig. 3.2: Mean equatorial zonal velocity profile (top), RMSE (middle), correlation (bottom) estimated over the period 1993 – 2015 for observations (TAO/TRITON: 0N-140W, 0N-110W,0N-165E,0N-147E, PIRATA:0N-23W,0N-0E and RAMA: 0N-90E.) (black) and GLORYS2V4 (green).

UV-CLASS2-MKE_1000m IV.3.3

Fig. 3.3: Mean value of the kinetic energy at 1000m from the ANDRO Argo drift data base (left) and GLORYS2V4 (right). Covered time period: 2002-2009.

Sea level IV.4

SL-CLASS1-TREND_2D IV.4.1





Fig. 4.1: SLA trend (mm/yr) from CMEMS altimetry SL TAC delayed time L4 product (left) and from GLORYS2V4 (right). Covered time period: 1993 – 2015.

SL-CLASS2-TIDE_GAUGE IV.4.2

Fig. 4.2: From left to right: RMSE (cm), correlation and explained variance (%) between the GLORYS2V4 reanalysis sea level and tide gauges observations (from GLOSS, BODC and SONEL networks).

SL-CLASS3-2DMEAN IV.4.3





Fig. 4.3: Domain averaged sea level monthly mean time series: Aviso SLA CCI observations (black), GLORYS2V4 (green). Covered time period: 19932011 for CCI data and 1993-2015 for GLORYS2V4.The bias between AVISO data and GLORYS2V4 is due to the use of different mean sea surface height and different period of the reference used to calculate the anomalies. Linear trends for SL-CCI and GLORYS2V4 are 2.6 mm/year and 3.1 mm/year, respectively.

SL-CLASS4 IV.4.4Not available yet for GLORYS2V4.

SL-DA-INNO_STAT IV.4.5

Fig. 4.4: Seal level innovation bias (left) and RMSE (right) over the whole domain (cm). Colours represent different altimetry satellites (ERS, T/P, Jason 1 & 2, Envisat, GFO, Saral/AltiKa, Cryosat2, HY2A). Covered time period: 1993 - 2012.





Transports IV.5

UV-CLASS3-VOL_TRANSP IV.5.1

Drake Passage

(A21)

Atlantic 11° S (A8)

Indian 32° S (I5)

Pacific 24° N (P3)

Pacific 32° S (P6)

South A. - Antarctica 30° E (I6)

Tasmania Antarctica

143° E (Sr3)

Lumpkin and Speer, 2007

129 ± 6 -0.5 ± 2.5 -13 ± 2.6 0.2 ± 2.6 14 ± 3 131 ± 8 141 ± 11

Ganachaud and Wunsh,

2000

140 ± 6 - -16 ± 8.77 1 ± 5.1 17 ± 6 - 157 ± 10

GLORYS2V4 159 ± 5 -1.6 ± 0.5 -17 ± 4 - 0.6 ± 0.6 19 ± 3 161± 5 178 ± 6

Tab 5.1: Mean volume transport across the seven WOCE sections in GLORYS2V4, and estimates provided by Lumpin and Speer (2007) and Ganachaud and Wunsch (2000) (in Sverdrups – 1Sv=106

m3.s-1).

Fig 5.1 Mean volume transport ( in Sv) of GLORYS2V4 over the 1993-2015 period and their interannual standard deviation (in red) compared to Lumpkin and Speer (2007)’s estimates and associated error (in blue).





UV-CLASS3-AMOC IV.5.2

Fig 5.2: Time series of maximum AMOC at 26°N in Sv from GLORYS2V4 in blue and from the RAPID array in red, plotted with a running mean of 10 days (thin line) and with a 3-month running mean (thick solid line). The time-mean values of the AMOC in the RAPID dataset is 17.4 Sv while it is 14.4 Sv for GLORYS2V4. The 10-days average time series between the ocean reanalysis and the observations are correlated at 0.86.





T-CLASS3-MHT IV.5.3

Fig 5.3: Meridional Heat Transport as a function of latitude (left: Atlantic, right: Global) (Petawatt). GLORYS2V4 (green) and estimates provided by Ganachaud and Wunsch (2000), Trenberth and Caron (2001) (black). Covered time period: 1993 - 2015.

Sea Ice IV.6

SI-CLASS1-CONC_MEAN IV.6.1Arctic





Fig 6.1.a: Arctic Sea Ice concentration climatology (in %) (1993-2015) for March (top), and September (bottom). From left to right: CERSAT observation and GLORYS2V4. Covered time period: 1993 - 2015. Values between 0 and 15% are not plotted.

Antarctic





Fig 6.1.b: Antarctic Sea Ice concentration climatology (in %) (1993-2015) for March (top), and September (bottom). From left to right: CERSAT observation and GLORYS2V4. Values between 0 and 15% are not plotted.

SI-CLASS1-CONC_EXTREME IV.6.2





Fig 6.2: Arctic Sea Ice concentration (in %) for September 1996 (top), and September 2007 (bottom). From left to right: CERSAT observation and GLORYS2V4. Values between 0 and 15% are not plotted.

SI-CLASS3-EXTENT IV.6.3





Fig 6.3: Integrated sea ice extent in the Arctic (left) and Antarctic (right) (units:106 km2). CERSAT observation (black) and GLORYS2V4 (green). Covered time period: 1993-2015.

SI-CLASS3-EXTENT_ANO IV.6.4

Fig 6.4: Integrated sea ice extent interannual anomalies in the Arctic (left) and Antarctic (right) (units:106 km2). CERSAT observation (black) and GLORYS2V4 (green). The linear correlation (not detrended) between GLORYS2V4 and CERSAT observation is provided in brackets for each hemisphere. Covered time period: 1993-2015.

SI-CLASS3-VOL IV.6.5





Fig 6.5: Integrated sea ice volume in the Arctic (left) and Antarctic (right) (units:106 km3) for GLORYS2V4 (green). Covered time period: 1993-2015.

SI-CLASS3-VOL_ANO IV.6.6

Fig 6.6: Integrated sea ice volume interannual anomalies in the Arctic (left) and Antarctic (right) (units:106 km3) for GLORYS2V4 (green). Covered time period: 1993-2015.

SI-CONC-DA-INNO-STAT IV.6.7





Fig. 6.7: Time evolution of the domain average innovation in terms of bias (left) and RMSE (right) (CERSAT data). Top: Arctic area, bottom: Antarctic area. Covered time period: 1993 – 2015.

Other IV.7

OTH-CLASS1-MLD_CLIM_SEASON IV.7.1





Fig. 7.1: Climatological mixed layer depth (m) in March (top) and September (bottom). From left to right: De Boyer Montegut climatology data and GLORYS2V4.

Documents

QUALITY INFORMATION DOCUMENT For Global Ocean …cmems-resources.cls.fr/documents/QUID/CMEMS-GLO... · QUALITY INFORMATION DOCUMENT For Global Ocean Reanalysis Products GLOBAL-REANALYSIS-PHY-001-025