GODAE is sponsored by
List of poster abstracts
[A-B] [C-E] [F-G] [H] [I-K] [L] [M-O] [P-R] [S] [T-W] [X-Z]
Authors P - R
Number 157 - Session 2
Three decades of monitoring of coastal oceanographic processes off southwest Indian coast
Paimpillil Sebastian Joseph1, M. Baba2, N.P.Kurian2, T.S. Shahul Hameed2
1 Center for Earth Research and Environment Management, K.K. Road, Cochin 17, India, 682017, Email psjoseph@eth.net
2Center for Earth Sciences Studies, Trivandrum, Kerala, India.
Abstract
The coastal oceanographic monitoring process had started nearly 3 decades ago with four indigenous CODAS (Coastal Oceanographic data acquisition system) spaced nearly at equal distance along the 560 Km stretch of southwest India coastline. The shallow and deep water wave data and its transformations to coastal region were utilized in the selection and design process of 4 medium fishing harbors. The deep water wave data from 2 sites along with the wave climate of the coastline was instrumental in the design and setting up of wave energy utilization plant at Vizhinjam and the unit had demonstrated the ability to generate 125KW electricity in the monsoon months. The deep water wave data was used in the modification and testing of Tohoku Hybrid Wave Predication Model and the model is currently in use for wave predictions of the region. The coastal erosion problem and mudbank formations were also studied with the wave energy distribution pattern along the coastal stretch. The implementation of coastal oceanography data for the welfare of the coastal population (ex. wave predications, coastal erosion warnings and prevention techniques and mudbank predictions) had encouraged the government authorities to continue is support to the project.
Number 159 - Session 5
Development of the data assimilation system for the Bering Sea
Panteleev G1., D. Nechaev, V. Luchin, P. Stabeno
1International Arctic Research Center, USA
Abstract
The conventional four-dimensional variational technique and quasi-stationary variational method are used to reconstruct climatologic hydrography and circulation of the Bering and Chukchi Seas. The Bering Sea circulation is reconstructed based on assimilation of water temperature and salinity, atmospheric data, Bering Strait water transport and approximately 500 surface drifters. Reconstructed water transports across Kamchatka and Near Straits are 28 Sv and 13 Sv, respectively.
The anomalies of sea surface heights (SSH) obtained from satellite altimetry and combined with the reconstructed SSH can be used for hindcasts of circulation in the vicinity of Aleutian Passes.
The important result of this study is estimate of the mean climatological sea surface height (SSH) that can be used as reference for satellite altimetry anomaly data in the Bering Sea region. Numerical experiments reveal that the obtained reference SSH combined with satellite altimetry observations (T/P and Jason-1) effectively constrains a realistic reconstruction of the Amukta Pass circulation.
Number 160 - Session 5
Reconstruction of the circulation in the Chukchi Sea
Panteleev G.
International Arctic Research Center, USA
Abstract
We present results of the reconstruction of the Chukchi Sea circulation during 1990-1991 and during the August-October 1995. The reconstruction of the ice drift and water circulation in the Chukchi Sea is based on the data assimilation system which combines a two ice-ocean models where sea ice data (drift and concentration) are assimilated employing nudging (optimal interpolation) and an ocean model with the traditional four-dimensional variational approach of oceanic data assimilation. The reconstructed circulation of the Chukchi Sea is in good agreement with observations. This data assimilation system has been used to determine optimal locations of moorings in the Bering Strait. Analysis of the reconstructed state reveals strong difference between summer and winter circulation. The anomalies of sea surface heights (SSH) obtained from satellite altimetry (ERS-2 and Envisat) and combined with the reconstructed SSH can be used for hindcasts of circulation in the Chukchi and East Siberian Sea.
Number 167 - Session 3
IMPROVED DESCRIPTION OF THE MESOSCALE VARIABILITY BY COMBINING FOUR ALTIMETER MISSIONS
Ananda Pascual1, Yannice Faugère2, Gilles Larnicol2 and Pierre-Yves Le Traon3
1Institut Mediterrani d'Estudis Avançats, IMEDEA(CSIC-UIB), Mallorca, Spain
2CLS, Space Oceanography Division, Toulouse, France
3 IFREMER, Brest, France
Abstract
Data of four satellite altimeters [Jason-1, ERS-2/ENVISAT, Topex/Poseidon interleaved with Jason-1 and Geosat Follow-On] are combined with the aim of improving the representation of the mesoscale variability in the global ocean. All the missions are intercalibrated and weekly gridded maps are produced. In areas of intense variability, the rms differences between a classical configuration of only two altimeters and the scenario merging four missions can reach up to 10 cm and 400 cm²/s², in SLA and EKE, respectively, which represents an important percentage of signal variance. In other areas of moderate activity, like the Mediterranean Sea, which is characterized by relatively small structures, the combination of Jason-1 + ERS-2/ENVISAT fails to reproduce some relevant mesoscale signals. On the contrary, when T/P and GFO are added, these features are well recovered and the EKE does not show significant discontinuities due to sampling effects. The merged Jason-1 + ERS-2/ENVISAT + T/P + GFO maps yield EKE levels 15% higher than Jason-1 + ERS-2/ENVISAT.
An external validation with surface drifters covering the global ocean is carried out, and it is shown that the four altimeter scenario resolves better some mesoscale structures that are not properly recovered with Jason-1 + ERS-2/ENVISAT. A comparison with 86 tide gauges reveals an improvement of about 25% in the estimation of sea level in coastal areas with 4 satellites compared to the errors obtained with 2 altimeters.
Number 109 - Session 4
INTENSIVE OBSERVATIONAL PROGRAMME USING ENVISAT AND GLIDER PLATFORMS TO MONITOR THE COASTAL OCEAN
A. Pascual1, S. Ruiz1, B. Garau1, Y. Faugère2, M. Martínez1, B. Casas1, M. Bonet1, G. Larnicol2, D. P.
Wang3, J. Tintoré1
1 IMEDEA(CSIC-UIB), Esporles, Mallorca, SPAIN.
2 CLS Space Oceanography Division, Toulouse, FRANCE
3 State University of New York, Stony Brook, US
Abstract
We present the results of an intensive observational programme conducted in the Balearic Sea combining new technologies (coastal gliders) and altimetry data from Envisat radar altimeter. The objective of this experiment is twofold: i) to investigate the limitations and potential improvements of altimetry data in the coastal area and ii) to test the feasibility of new technologies to study coastal dynamics.
The strategy consisted of repeated samplings along Envisat 773 track, which is perpendicular to the Balearic current (one of the main oceanographic features of the Balearic Sea). The samplings spanned from July 2007 to June 2008 (one year time series) and they were carried out periodically every 70 days, always simultaneously with the Envisat passage.
Preliminary results show reasonable coherence between absolute dynamic topography from altimetry and the dynamic height from the glider CTD data, although altimetry observations near the coast are missing in some cases. This puts in evidence the need for specific altimetric algorithms (retracking, corrections, editing, etc) for coastal areas.
We then focus on one particular mission (April 2008) revealing the appearance of an anomalous anticyclonic eddy detected by altimetry, glider, SST and model data. A methodology is developed to combine surface glider geostrophic velocities with integrated currents estimated by the glider (GPS locations every 6 hours), which proves to be very efficient for reducing the differences between altimetry and glider data. Some hypotheses for the formation of this anomalous structure will be also explored.
Thus, the potential of new autonomous platforms in combination with satellite data to investigate the coastal ocean dynamics is promising. Further, these studies provide interesting insights of the present limitations and possible future improvements of coastal altimetry.
Future work include reproducing this exercise along Jason-1/2 tracks in the frame of OSTST project as well as applying improved altimeter corrections (tidal model, mean profile, MOG2D HR, troposphere correction), high frequency sampling data and reviewing the data recovery near coast in the frame of the PISTACH project.
Number 168 - Session 4
ON THE QUALITY OF REAL-TIME ALTIMETER GRIDDED FIELDS: COMPARISON WITH IN SITU DATA
Ananda Pascual1, Christine Boone2, Gilles Larnicol2, Pierre-Yves Le Traon3
1 Institut Mediterrani d'Estudis Avançats (CSIC-UIB), Mallorca, Spain
2 CLS Space Oceanography Division, Toulouse, France
3 IFREMER, Brest, France
Abstract
The timeliness of satellite altimeter measurements has a significant impact on their value for operational oceanography. In this paper, we use an OSE (observing system experiment) approach to assess the quality of fast delivery (FD) altimeter products, a key issue for a robust monitoring and forecasting of the ocean state. In addition, the impact of two improved geophysical corrections and the number of missions that are combined in the altimeter products are also analyzed. The improved tidal and atmospheric corrections have a significant effect in coastal areas (0-100 km from the shore), and a comparison with tide gauge observations shows a slightly better agreement with the gridded delayed-time Sea Level Anomalies (SLA) with two altimeters (Jason-1 + ERS/ENVISAT) using the new geophysical corrections (mean square differences in percent of tide gauge variance of 35.3 %) than those with four missions (Jason-1 + ERS/ENVISAT + TOPEX/POSEIDONinterlaced + Geosat Follow-On) but using the old corrections (36.7 %). In the deep ocean, however, the correction improvements have little impact. The performance of FD products versus delayed-time data are compared using independent in-situ data (tide gauge and drifter data). It clearly highlights the degradation of real-time SLA maps versus the delayed-time SLA maps: Four altimeters are needed in real-time to get the similar quality performance as two altimeters in delayed-time (sea level error misfit around 36% and zonal and meridional velocity estimations errors of 27 % and 33%, respectively). This study proves that the continuous improvement of geophysical corrections is very important, and that it is essential to stay above a minimum threshold of four available altimetric missions in order to capture the main space and time oceanic scales in FD products.
Number 103 - Session 4
IN-SITU DELAYED MODE AT CORIOLIS DATA CENTER
C.Pertuisot, E.Brion, C.Coatanoan, F.Gaillard, T.Carval, L.Petit de la Villeon
IFREMER, Plouzane, France
Abstract
End of 2007, the Coriolis Data Center has set up a new product dedicated to operational oceanographic centers that want to perform re-analysis on a delayed mode basis. The release 2007 covers the period 2002 - 2006, the next one (end 2008) is extended to 1990-2007.
In addition to the near real time validation done on a daily and weekly basis for the forecasting needs, it has been decided to create a reference dataset updated on a yearly basis. The new procedure has involved an objective analysis method (statistical tests) with a visual quality control (QC) on the suspicious profiles, and has been developed to improve the database content and to fit the level required by the physical ocean reanalysis activities.
The quality control process uses two runs of objective analysis, corresponding to two different time windows, with an additional visual control in between. The first run is done on a three weeks window to capture the most doubtful profiles which are visually checked by an operator to decide whether or not they are bad data or real oceanic phenomena. Whereas the second run is operated on a weekly basis for the modeling needs.
The reprocessing of the 2002-2006 period is global and annual delayed analysis of the content of the database and an additional validation of the dataset collected in real time and delayed mode during this 5 years period. The release provides T and S weekly gridded fields and individual profiles both on their original level and interpolated level.
This new Coriolis product is available on different servers using different technologies (ftp, OPeNDAP and web).
http://www.coriolis.eu.org/cdc/global_dataset_release_2007.htm
http://www.coriolis.eu.org/cdc/global_dataset_release_2008.htm
Number 56 - Session 5
Seasonal surface circulation of south Indian Ocean derived by Combining Drifter and Satellite observations during 1994 - 2004
Benny Peter
Abstract
Eulerian mean velocity field of south Indian Ocean is estimated with a resolution of 1/3 x 1/3 degree in latitude and longitude by combining satellite altimeter and surface drifting buoy observations. High resolution Maps of sea Level Anomaly (merged T/P or JASON and ERS1/2 or Envisat ) and the surface drifter data from the Global Drifter program during 1993 January -2004 January have been employed. Mean Winds Fields measured by scatterometers onboard ERS/QuikSat is used for computing the wind-driven Ekman current. The estimated mean velocity field illustrates strong boundary currents, zonal currents; weak flows prevail at the central part. The Agulhas Current velocity even reaches 2m/s at around 35° S. The distribution of energy associated with the fluctuating motion and the mean flow illustrates that mesoscale variability are particularly relevant in the Mozambique Channel and around 40°S. The estimated eddy fluxes confirm the advection of mesoscale features along the western boundary. The seasonal averages of instantaneous velocity field of different years are determined and analyzed for understanding the long-term variability occurred in circulation.
The long-term average monthly surface circulation pattern illustrates large variations in the flow field. Among the surface currents, the South Equatorial Current is subjected to latitudinal shift as well changes in its structure and characteristics.; The South Equatorial Current is strong, broad and meandering in southern summer period, whereas it is weak and narrowed in winter. In western region the East Madagascar Current and Agulhas Current shows slight seasonal changes. In winter, the circulation is weak compared to summer. In the eastern boundary the Leeuwin Current is well developed in January ?April, but is weakened in July. In October even a reverse flow is initiating between 26°S and 28°S.
Number 99 - Session 4
PROVIDING AN OCEAN IN SITU DATA SERVICE FOR THE NEEDS OF
OPERATIONAL OCEANOGRAPHY
L.Petit de la Villéon, S. Pouliquen, C. Coatanoan, T. Carval, V. Bernard, E. Brion, C.Pertuisot
Ifremer, Plouzane, France
Abstract
Seven French research agencies involved in ocean research and ocean predictions are together developing a strong capability in operational oceanography based on three components including altimetry (Jason), digital modelling with assimilation (Mercator) and in situ data service (Coriolis).
The Coriolis data centre aims to collect, quality control and distribute ocean data worldwide in both near real-time and delayed formats for assimilation and validation purposes. Furthermore, the Coriolis data centre is able to deliver products such as T & S fields and reference datasets.
To be able to deliver such a global dataset, the Coriolis data centre plays an important role in three major JCOMM projects:
- The Argo project, where the Coriolis data centre acts as one of the two global data centres. The comprehensive Argo dataset is available via the Coriolis server. This dataset holds the data from 3111 active profiling floats (June 24th 2008).
- The GOSUD project, which aims to process and distribute sea surface data collected by both research vessels and merchant ships when they are at sea. For the moment, only SST & SSS surface data are taken into account but the objective is to extend the project to include other parameters such as oxygen, fluorescence or PCO². The Coriolis data centre is one of two GOSUD global data centres. Since the beginning of 2008, forty ships have sent surface data to the Coriolis data centre.
- The OCEANSITES project which collects and processes data from deep open ocean time series sites. Data from 60 different platform deep sea moorings are available at the Coriolis data centre which acts as one of the two global data centres.
In order to complement this dataset collected within the projects mentioned above, connections to the GTS - the WMO network for data exchanges - have been implemented to retrieve any of the ocean data which were not part of the three projects Argo, Gosud and OceanSites.
All the datasets described above are freely distributed on a daily or regular basis via different servers using different technologies (ftp, OPENDAP, Thredds and web).
From this complete dataset, value added products are produced and delivered regularly. The reprocessing of the 2002-2006 period produced a global data set better validated that provides three products: T & S gridded fields, individual quality controlled profiles both at original and interpolated levels. We will reprocess the 1990-2007 dataset before the end of 2008.
Number 169 - Session 4
THE ODYSSEA GLOBAL SEA SURFACE TEMPERATURE ANALYSIS FOR EUROPEAN OPERATIONAL OCEANOGRAPHY
J.F.Piollé1, E.Autret1, H.Roquet2, P.Le Borgne2, F.Orain2
1Ifremer, Plouzané, France
2Météo-France, Lannion, France
Abstract
In the frame of the European project MERSEA for the implementation of an operational oceanography capability in Europe, Ifremer has set up the ODYSSEA analysis system producing the global high resolution sea surface temperature fields required by the various ocean models and downstream services, following a methodology jointly defined at European level with Météo-France, Met.No, DMI and CNR. These fields are being produced daily on a 0.1 degree grid since January 2007. They are estimated by an optimal interpolation method merging SST satellite measurements from both infrared and microwave sensors. The processing scheme is broken down in two main steps.
The first step is the collection and preprocessing of all available GHRSST L2P products (except MODIS for the current version). The pre-processing consists mainly in a screening and quality control of the retrieved observations from each single datasets and in constructing a coherent merged multi-sensor set of the most relevant and accurate observations. The merging of these observations requires a method for bias estimate and correction (relative to a chosen reference, currently AATSR). This inter-sensor correction is daily updated using the latest data and each single source dataset monitored through a set of statistical tools allowing to detect any problem with a source and to continuously upgrade and improve the screening of the input data. The gap free SST foundation field is finally computed from the merged set of selected observations using an objective analysis method. Here again the performances and properties of the MERSEA analysis are estimated and monitored both daily and on a long term basis through a set of evaluation tools. Particular attention is paid to assessing the true product resolution, correct retrieving of the main oceanic features, accuracy, long-term consistency. Appropriate methods were implemented for the overall assessment of the product quality with respect to each of these criteria.
Number 138 - Session 3
DECADAL VARIABILITY IN REGIONAL AND GLOBAL MEAN SEA LEVEL AND ITS CAUSES
R. M. Ponte1, C. Wunsch2, P. Heimbach2
1Atmospheric and Environmental Research, Inc., Lexington, MA, USA
2Massachusetts Institute of Technology, Cambridge, USA
Abstract
The determination of regional and global mean sea level change and its causes is a problem of great societal interest, given the potential impact of sea level rise on many densely populated coastal regions. Unfortunately, no published estimate is a true global mean, given the poor data coverage particularly below several hundred meters and in the Southern Ocean, and claimed uncertainties do not account for the likely presence of systematic errors in the data. Similar issues affect the study of regional sea level patterns.
A comprehensive attempt to deal with the sea level problem in a climate context is one focus of the Estimating the Circulation and Climate of the Ocean (ECCO) project and its current ECCO-GODAE extension. As part of ECCO-GODAE, 3-dimensional estimates of the time-dependent ocean state are produced by fitting a general circulation model to most available ocean data in a constrained least-squares optimization procedure. For the period analyzed (1992-present), several hundred million ocean observations are used, including all the altimetry and modern hydrography, Argo floats, sea surface temperature, and more recently estimates of the mean oceanic mass provided by space-based gravity measurements. A major effort was devoted to assessing the errors in all data, and which often differ from those commonly accepted. The ECCO-GODAE solutions are dynamically and kinematically consistent with all surface boundary conditions and model equations, and within the expected error bars of model and data; methodology and other details are described in an accompanying paper by Heimbach et al.
Regional sea level patterns exhibit decadal variability with magnitudes substantially larger than expected long-term trends in global mean sea level (GMSL) and are associated with oceanic circulation patterns involving both steric and mass changes, with important contributions from temperature and salinity effects throughout the water column, including the abyss. Significant quantitative differences exist between the combined estimates and those based upon partial use of the data base. Quantitative attribution of GMSL changes to heating and freshwater addition, even in the data-rich past 15 years, is still very difficult. In this light, we discuss in detail various possible implementations of data constraints on mean quantities such as global ocean mass, heat content and other relevant variables for GMSL, as well as related model issues on how to best formulate surface freshwater boundary conditions and deal with the volume conservation implicit in the Boussinesq approximation.
Number 146 - Session 1
IMPLEMENTATION OF THE OPERATIONAL OCEANOGRAPHY SYSTEM IN THE IBERIA-BISCAY-IRLAND AREA (IBI-ROOS)
S. Pouliquen1 Alicia Lavin2 and IBI-ROOS Consortium
1 IFREMER Brest France
2 IEO Santander Spain
Abstract
A large number of human sea-related activities as fishing, maritime transport, aquaculture, security and rescue, toxic events warming are of high economic and social importance in the Iberia-Biscay-Ireland area and were the main motivation for the establishing of an Operational Oceanography System in the area Iberia- Biscay-Ireland. IBI-ROOS region comprises almost the Celtic Sea the whole of the Bay of Biscay the western Iberian margin and the Gulf of Cadiz. In 2003, institutes from the five countries surrounding the area decided to set up the IBI-ROOS consortium in partnership with the international communities and agencies; it constitutes a close co-operation between national governmental marine laboratories and agencies in the countries surrounding the area responsible for collection of observations, model operations and production of forecasts services and information for the industry the public and other end users.
Beginning of 2008, 12 institutes had signed the Memorandum of Understanding and 6 were working on it. The first version of the Implementation Plan was published by EuroGOOS at the end of 2006. IBI-ROOS organization was designed during the first meeting in February 2007 and a portal was developed (http://www.ibi-roos.eu) where information on the IBI-ROOS Observing System are provided as well as first integrated access to data.
The two main issues the consortium decided to work on in 2008-2009 are to improve access to observation data, setting up data access portal, and downscaling from regional to coastal modeling. These activities are developing with support of EU projects such as ECOOP and MyOcean and with national funding. Networking between the team is a clear factor of success.
Many marine related industries and services can substantially benefit from an ocean observing system based on the operational oceanography concept. The observing and forecasting of harmful algae blooms, fish stocks, pollution, water quality, sea level and waves are of a great importance not only to the potential users whose activities are carried out in the IBI-Roos region but also to a larger Atlantic Community.
Number 144 - Session 4
DESIGNING COHERENT NETCDF FORMAT FOR JCOMM IN-SITU NETWORKS
T. Carval, S. Pouliquen and Argo, Gosud, OceanSITES Data Management Committees
Ifremer, Brest, France
Abstract
In 2001, when we started designing the Argo Data management architecture for profiling float data, there were not any on the shelf framework that we could have used to encode Argo data. Nonetheless Netcdf was emerging with a variety of scientific software able to manipulate them. BODC/ GF3 vocabulary was meeting our requirement for temperature and salinity and we agreed pretty rapidly on a common vocabulary/units and set of attributes. When in 2003, within GOSUD, for underway data acquired from vessels, we had to address the same issue, with a bit more variable, it seemed clear to the GOSUD data management team that format and standard used needed to be coherent with Argo as the parameters measured and used jointly by the operational oceanography community as well as the research. Finally in 2005 the same issue was raised within the OceanSites data management team for time series reference sites. The problem was then a bit more complicated as the number of parameters was more important and was mixing both ocean and atmospheric variables, profile and time series data as well as multidimensional data such as current from ADCP. We decided then to approach the CF community and together with them we extended the Argo/Gosud format to meet OceanSites requirements and identified actions to upgrade Argo and Gosud format to be CF compliant in order to benefit from the tools designed for CF compliant files. Moreover within the EU FP6 SeaDataNet project collaboration started between CF and GF3 community to converge of vocabularies and implement gateways between the two vocabularies.
Therefore we have presently with OceanSites format a framework able to handle other oceanographic dataset that is pretty handy for users. We plan to use this format within the MyOcean project to integrate in-situ data at European level.
Argo: http://www.coriolis.eu.org/cdc/argo_rfc.htm
GOSUD: http://www.ifremer.fr/gosud/documentation.htm
OceanSites: http://www.oceansites.org/data/index.html
Number 18 - Session 4
THE KUROSHIO EXTENSION NORTHERN RECIRCULATION GYRE: PROFILING FLOAT MEASUREMENTS AND FORCING MECHANISM
B. Qiu1, S. Chen1, P. Hacker1, N. Hogg2, S. Jayne2, H. Sasaki3, ...
1University of Hawaii, Honolulu, HI, USA
2Woods Hole Oceanographic Institution, Woods Hole, MA, USA
3Earth Simulator Center, JAMSTEC, Yokohama, Japan
Abstract
Mid-depth, time-mean circulation in the western North Pacific Ocean (28°N-45°N, 140°E-165°E) is investigated using drift information from the profiling floats deployed in the Kuroshio Extension System Study (KESS) and the International Argo programs. A well-defined, cyclonic recirculation gyre (RG) is found to exist north of the Kuroshio Extension jet, confined zonally between the Japan Trench (~145°E) and the Shatsky Rise (~156°E), and bordered to the north by the Subarctic Boundary along ~40N. This northern RG, which is simulated favorably in the eddy-resolving OFES hindcast run model, has a maximum volume transport at 26.4 Sv across 159°E and its presence persists on the interannual and longer time scales. An examination of the time-mean x-momentum balance from the OFES hindcast run output reveals, that horizontal convergence of Reynolds stresses works to accelerate both the eastward-flowing Kuroshio Extension jet and a westward mean flow north of the meandering jet. The fact that the northern RG is eddy-driven is further confirmed by examining the turbulent Sverdrup balance, in which convergent eddy potential vorticity fluxes are found to induce the cyclonic RG across the background potential vorticity gradient field.
For the strength of the simulated northern RG, we find the eddy dissipation effect to be important as well.
For more information, visit http://www.soest.hawaii.edu/oceanography/bo/NRG_JPO.pdf
Number 28 - Session 3
GULF STREAM PATHWAY AND TRANSPORT VARIATIONS:
OBSERVATIONS AND MODELING
J. G. RICHMAN
Naval Research Laboratory, Stennis Space Center, MS, USA
Abstract
The Gulf Stream separates from the coast of North America at Cape Hatteras and flows into the North Atlantic as a narrow meandering jet. As the jet meanders, rings are pinched off to the north (warm core) and south (cold core). Direct current observations (Hogg, 1992) suggest that the transport of the Gulf Stream increases markedly downstream, from approximately 95 Sv near Cape Hatteras to 150 Sv east of the New England Seamount Chain (~65°W, due to an increase primarily in the barotropic flow. Eddy resolving models (1/10° POP, Smith, et al. 2000, and 1/12° HYCOM) suggest a more complicated picture. In the models, the mean transport increases from Cape Hatteras to the New England Seamount Chain near 65°W, decreases by more than 50 Sv at 60°W and then increases again to nearly 150 Sv west of the Newfoundland Rise.
To investigate the path and transport variability of the Gulf Stream and the North Atlantic Current, we have extracted the sea level, sea surface temperature and transport from 5 years (2003 ? 2007) of the global 1/12° HYCOM ocean simulation run at NRL with and without data assimilation. The statistics of the sea surface height variability from the model are compared to the 14 years of altimetric sea levels. The path and transport of the Gulf Stream are estimated from the altimetric sea level using a modification of the Gaussian jet model technique of Kelly and Gille (1990). Our implementation of the technique allows for multiple crossing of the Gulf Stream as well as the existence of rings. Using this technique, we can determine the path and transport of the Gulf Stream every 10 days. For consistency, the same technique is applied to the model sea level variations. In addition, the path of the Gulf Stream is estimated from AVHRR imagery, often designated as the north wall of the Gulf Stream. The mean path of the Gulf Stream is similar between the model, altimetric and IR frontal observations, although the model paths are south of the altimteric path and the IR fronts are consistently north of the altimetric path. Assuming that the transport increases are primarily barotropic as suggested by Johns, et al. (1995), the altimeter and models show similar results for the mean transport of the Gulf Stream with the transport increaseing to the New England Seamount Chain (~65°W) , decreasing by 50 Sv to 60°W and then increasing again until just west of the Newfoundland Rise. At the Newfoundland Rise, the Gulf Stream bifurcates into the North Atlantic Current and the Azores Current. However, the models tend to underestimate the total sea level difference across the Gulf Stream by approximately 20 cm over the region.
Number 154 - Session 3
USING DYNAMIC TOPOGRAPHY TO ASSESS THE QUASI STATIONARY CIRCULATION IN THE NORTH ATLANTIC
F. Richter1, D.Sidorenko1, H. Reinhardt1, J. Schröter1
1Alfred-Wegener-Institut, Bremerhaven, Germany
Abstract
The data available from Argo profiling buoys and satellite altimetry are combined into Inverse Finite Element Ocean model for each of the years 2003 to 2006. The model solves for temperature and salinity fields that are close to Argo, respect quasi-stationary dynamical balances, and simultaneously produces estimates of the circulation. The experiments with and without MADT (seal level anomaly + Rio05) included into IFEOM as a weak constraint have been carried out. Variability of the North Atlantic Ocean circulation is further compared to the tide gauge measurements. We show that including altimetry improves the circulation patterns.
Number 82 - Session 4
A MEAN OCEAN DYNAMIC TOPOGRAPHY DERIVED FROM ARGO FLOATS, SATELLITE ALTIMETRY AND AN OCEAN MODEL
K. R. Ridgway, J. R. Dunn, P. Oke, R. Fiedler...
Centre for Australian Climate and Weather Research, CSIRO, Hobart, Australia
Abstract
We present an improved estimate of the absolute mean ocean dynamic topography (MDT) for the period 1992-2007 and covering the region [100°E-160°W; 50°S - 10°N]. Underpinning this estimate is an enhanced high-resolution ocean climatology (T, S) for the same region. Satellite altimetry observations are used to overcome the inadequacies of the sampling of the in situ data. Data profiles collected by the Argo network commencing in 1999 and approaching full strength in 2003, provide a huge improvement in spatial and temporal sampling over that obtained from the traditional source of CTD transects occupied during research cruises. However, the research cruise data are still essential for coastal regions. The distribution of Argo profiles enables an accurate description of the mean ocean properties at large scales to be obtained.However, even with this dataset, averaging is required which tends to smear out mesoscale features. We first use the very high spatial and temporal sampling of the satellite altimetry to adjust the mean surface steric height field and address the gaps in the spatial and temporal distributions of the in situ data. The balance between baroclinic and barotropic contributions to the surface height topography is demonstrated. The surface height correction is then projected through the water column into the T and S fields using an empirical model derived from the historical data archive. Improvement from eddy to basin scale is shown in the resultant fields. The corrected mean T and S fields are assimilated into a high resolution ocean model (Ocean Forecasting Australia Model, OFAM). In a 2-year run, OFAM is forced by seasonal winds and strongly nudged to the climatological fields. The resulting absolute mean topography agrees closely with existing observationally based estimates at large spatial scales while retaining the realistic mesoscale structure derived from the assimilated T and S. The new MDT was developed within the BLUElink project and will be used in future hindcast runs of OFAM.
Number 123 - Session 5
The GMAO Ensemble Kalman Filter and its use in Seasonal Forecasts
Guillaume Verniers1, 2 Michele R. Rienecker1, Robin M. Kovach1, 2 and Christian L. Keppenne1,2
1GMAO, NASA/Goddard Space Flight Center, Greenbelt, Maryland, USA
2SAIC inc., Beltsville,Maryland, USA
Abstract
A multivariate ensemble Kalman filter (EnKF) is now used routinely in the GMAO ocean data assimilation system (ODAS). The EnKF assimilates in situ temperature (T) and salinity (S) data and remotely sensed sea surface height (SSH) observations into the Poseidon v4 global quasi-isopycnal ocean model. The EnKF updates the model's T, S and currents and the layer thicknesses adjust gradually to the updated T and S by means of the state equation. An online bias correction algorithm continuously estimates the sea surface height bias and the SSH innovations are adjusted accordingly before being assimilated.
Several continuous assimilations have been conducted to investigate the impacts of specific data types, i.e., specifically Argo S and SSH data, for seasonal forecasting applications. At the end of each month, each member of the ensemble of ocean state estimates provided by the EnKF is coupled to the GMAO atmospheric general circulation model and the resulting coupled-model ensemble is run in forecast mode for one year. The ensembles of seasonal-hindcast time series thus obtained are compared to observations and to seasonal hindcasts obtained with a coupled-model ensemble in which the ocean initial conditions are obtained by perturbing the GMAO univariate ocean analyses. The latter analyses rely on an optimal interpolation (OI) algorithm to assimilate the T and S observations and to update the model T and S fields.The univariate analyses also assimilate synthetic salinity observations assembled for the in situ TAO, XBT and PIRATA T measurements, but do not assimilate SSH.
The impact on the seasonal predictions of assimilating the Argo S observations and TOPEX/Jason SSH observations during the forecast initialization phase is assessed by examining hindcasts of SSH, and of ocean heat content and salt content. The results show that the multivariate EnKF generally outperforms the univariate OI algorithm and that Argo and altimetry data seem to work in tandem to improve the upper ocean forecasts. The impact of SSH by itself is difficult to assess because of forecast drifts, however, the SSH does improve the forecasts of upper ocean heat content from July starts when predictability has a longer time scale.
Number 89 - Session 5
The ENSEMBLES ocean reanalysis and forecast database for seasonal to decadal climate variability and predictability studies
P. Rogel, F-J Doblas-Reyes, and the seasonal to decadal ENSEMBLES partners
Abstract
A unique ocean data base has been constituted for the seasonal to decadal ENSEMBLES experiments. This data base gathers outputs from an ensemble of data assimilation systems and coupled hindcast systems, on a single storage facility at ECMWF. A common set of variables have been stored with a common Netcdf format, interpolated on a common 1°x1° resolution, 33 vertical levels grid, and made publicly available. This set of variables includes temperature, salinity, velocity (along the 3 components), upper ocean heat content, sea level height, mixed layer depth and depth of the 20°C isotherm. Data are available at http://ensembles.ecmwf.int/thredds/catalog/ocean/catalog.html
Data from two different origins have been stored using these conventions. The first one is an ensemble of ocean reanalyses over the ERA40 period (1960-2006). These reanalyses almost all use classically ERA40 data as forcing, the EN3 quality-controlled temperature and salinity profiles data base built at the UK Met-Office, and have been strongly constrained by the observed Sea Surface Temperature. Reanalyses from several groups (including ECMWF, INGV, UK Metoffice, IfM and CERFACS) are stored and comparison of a few key diagnostics will be showed.
The second set of ocean data consists of ocean hindcasts generated from the different seasonal to decadal experiments produced for the project, and especially the Stream 2 experiments which cover the whole ERA40 period. Data stored include seasonal (up to 7 months, every season), annual (up to 14 months every year) and decadal (up to 10 years every five year) retrospective forecasts of the aforementioned groups plus Meteo-France. A preliminary assessment of those hindcasts will also be showed.
(Last Updated: 30-10-2008)