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 S
Number 171 - Session 2
DOES SMOS ACCOMPLISH GODAE REQUIREMENTS? ISSUES CONCERNING SATELLITE SALINITY RETRIEVAL
R. Sabia1,2, A. Camps1,2, M. Talone1,2, M. Vall-llossera1,2, J. Font2,3
1Universitat Politècnica de Catalunya, Remote Sensing Laboratory, Barcelona, Spain
2SMOS Barcelona Expert Centre on Radiometric Calibration and Ocean Salinity, Barcelona, Spain
3Institut de Ciències del Mar, Dept. d'Oceanografia Física, Barcelona, Spain
Abstract
ESA's Soil Moisture and Ocean Salinity (SMOS) mission will be launched in early 2009 to provide sea surface salinity (SSS) maps on a global basis with frequent temporal coverage. The SMOS ocean community is currently defining a reliable inversion scheme to obtain SSS estimations from the multi-angular brightness temperatures (TB) provided by the MIRAS (Microwave Interferometric Radiometer by Aperture Synthesis) sensor.
The salinity retrieval issues whose influence is critical in the inversion procedure are: 1) Scene-dependent bias in the simulated TBs, 2) Radiometric sensitivity (thermal noise) and radiometric accuracy, 3) L-band forward geophysical model function definition, 4) Auxiliary data uncertainties, 5) Restrictions in the cost function, especially on salinity term, and 6) Adequate L3 spatio-temporal averaging.
Different tuning and setting of the minimization algorithm lead to different results, and complete awareness of that should be assumed. To support this, a comprehensive salinity error budget analysis has been performed to evaluate the extent of the impact of different variables and parameterizations in the salinity error. An error budget gives a whole perspective of the magnitude of each single error contribution and is suitable to furnish a sketch of the problems that will have to be addressed and tackled once real SMOS data will be down-linked .
An important characteristic of this study is that the simulations have been carried out using the SMOS End-to-end Performance Simulator (SEPS), an ad hoc tool that takes into account all the instrument specific features. TB maps generated by SEPS have the realistic features induced by the image reconstruction algorithm such as biases and the pixel-dependent radiometric accuracy. A customized ocean salinity processor, in turn, is in charge of producing SSS maps out of SEPS TBs.
Issues like radiometric sensitivity, radiometric accuracy, Faraday effect, Sun contamination and some extra analysis on the variations introduced by using a different seawater dielectric constant model or including a foam parameterization have been studied. Besides, several efforts have been made in quantitatively assessing the impact of different auxiliary datasets (sea surface temperature and wind speed) on the retrieved SSS accuracy.
The salinity error is computed after spatio-temporal averaging, to be consistent with the GODAE accuracy requirements: 0.1 psu, every 30 days and 100x100 km2. This study indicates that, when considering the uncertainties in the sea state descriptor, the total salinity error will exceed the mission requirements, by a rate which is variable according to the configuration chosen (polarization, restrictions, knowledge of the a priori auxiliary fields, and so on).
After launch, nevertheless, the algorithm will surely need refinements, either way in the forward model and in finding a closed formulation of the cost function. In fact, once the real data will be available, it will be possible to tailor properly an adequate model to the SMOS data; to this end, a semi-empirical formulation including several sea state descriptors will have to be adjusted to the down-linked TBs with the aim of improving the salinity retrieval.
Likewise, a key issue is a proper balancing of the different terms of the cost function. With respect to the background terms, a sensitivity study will ensure that the different constraining terms are homogeneously minimized. A balancing of the cost function will be achieved by introducing empirical weights to each single term of the function. Bias mitigation is still an issue, and further efforts will have to be devoted to this topic. Eventually, advanced optimization techniques will be used to properly average the single-overpass SMOS data into the GODAE-like boxes.
Number 39 - Session 4
NON-EQUIVALENCE OF PARALLEL AND SEQUENTIAL ASSIMILATION IN PRESENCE OF LOCALISATION
P. Sakov, L. Bertino
1Nansen Environmental and Remote Sensing Center, Bergen, Norway
Abstract
Localisation in the ensemble-based data assimilation systems is an essential and widely used technique for increasing the effective rank of the ensemble. While it is known that in the Kalman filter framework sequential and parallel assimilation of observations with non-correlated errors are equivalent, we are not aware of studies on whether this still holds for data assimilating systems with localisation. At the same time, some data assimilation schemes, such as Potter's scheme for Square Root filter, are based on the assumption of equivalence of sequential and parallel assimilation.
This work draws attention to the non-equivalence of parallel and sequential assimilation in the ensemble Kalman filter framework when using localisation and to its possible practical implications.
Number 177 - Session 3
PREDICTION OF INDIAN OCEAN DIPOLE USING ALTIMETRY AND IN-SITU MEASUREMENTS
I. V. Sakova1,2,3, R. Coleman2,3,4, G. Meyers5
1 NERSC,Bergen,Norway
2 CSIRO Marine Research, Hobart, Australia
3 School of Geography & Environmental Studies, University of Tasmania,, Hobart, Tasmania, Australia
4 Antarctic Climate & Ecosystems CRC,, Tasmania, Australia
5 Integrated Marine Observing System, University of Tasmania,Australia
Abstract
Indian Ocean Dipole (IOD) is the major climate phenomenon that influences the climate of the whole Indian Ocean region. Prediction of IOD is important for reduction of its impact on the economies of the rim countries. While the IOD was first described in 1999, its predictability is still an open question, for two reasons. Firstly, the driving mechanism of the IOD is not yet fully understood; and secondly, the signal itself is not as strong as ENSO and is therefore harder to separate from other signals.
In our study we have identified the link between the IOD events in the past and the amplitude of the 18-month signal in the Indian Ocean. We argue that it is possible to predict the development of IOD from the observed dynamics of the 18-month signal from altimetry and in-situ measurements. These measurements make it possible to identify the earlier changes in the depth of the thermocline rather than later changes of SST.
On example of the recent 2006-2007 IOD we show that the associated development of the 18-month signal could be identified from observations as early as in 2004. We argue that additional in-situ measurements close to the Sumatra-Java coast could be very beneficial for the IOD prediction.
Number 60 - Session 4
THE UPPER CELL OF THE SOUTHERN OCEAN
Jean-Baptiste Sallee
Centre for Australian Climate and Weather Research, CSIRO, Hobart, Australia
Abstract
The intensity and location of the upper overturning cell in the Southern Ocean is uncertain, for instance if the cell is associated with subduction centered on the Subantarctic Front or the Polar Front. Argo data provide a new view of the Southern Ocean interior and allow for the first time to resolve the seasonal cycle of the upper ocean. This new dataset is used to revisit the role of the mean flow in subducting water masses in the vicinity of the Antarctic Circumpolar Current. New eddy diffusivity estimates are also used to revise the main balances of the 2-d overturning cell, and it is found that horizontal buoyancy mixing plays an important role near the main fronts of the ACC. The new estimate of kappa allows us to also revisit the estimate of the eddy induced advection through the parameterization of Gent and McWilliam (1990). Below the mixed layer, we find a convergence of the 2-d upper cell centered on the Polar Front. However, we have also shown that one can get a large range of results when working within the 2-d framework. Eddy diffusivity in the Southern Ocean has been estimated from several approaches in order to quantify the role of eddies on mixed layer heat and mass budgets, and to revisit the role of eddies on the upper cell of the meridional overturning circulation. Estimates were made with surface drifters. The pattern and the order of magnitude of kappa is compared to other studies. We consider the upper cell in a 3-d framework, unveiling strong inhomogeneities of the upper cell convergence. We estimate a proxy of the convergence by computing a budget of transport in the mixed layer based on observational data and mixed layer parameterizations.
Number 13 - Session 5
ON THE USE OF GODAE AND SATELLITE PRODUCTS TO IMPROVE COASTAL SIMULATIONS ON THE NORTHERN GULF OF MEXICO
R. V. Schiller1, V. H. Kourafalou1, P. J. Hogan2, O. M. Smedstad3, G. Halliwell1 and G. Goni4
1University of Miami - Rosenstiel School of Marine and Atmospheric Science, Miami, USA
2Naval Research Laboratory - Stennis Space Center, Stennis, USA
3QinetiQ North America, Technology Solutions Group - PSI, Stennis Space Center, MS, U.S.A.
4National Oceanic and Atmospheric Agency - Atlantic Oceanographic and Metereological Laboratory, Miami, USA
Abstract
The coastal and shelf areas in the Northern Gulf of Mexico (NGoM) are subject to intense interactions with offshore flows, namely the Loop Current (LC), the Loop Current warm core eddies (LCEs) and the Loop Current frontal eddies (LCFEs). A nesting modeling approach is thus necessary to represent coastal to offshore interactions. Furthermore, the realistic representation of the extension of the Loop Current (and hence its proximity to the NGoM), as well as the position and size of the eddies approaching the shelfbreak require boundary conditions from a data assimilative regional model.
A high resolution (1/50°) model has been developed for the Northern Gulf of Mexico region, based on the Hybrid Coordinate Ocean Model (HYCOM) and hence called the NGoM-HYCOM model. It is nested within the regional, data assimilative Gulf of Mexico GoM-HYCOM model (1/25° resolution), which employs GODAE products for boundary conditions. The study objective is to simulate the NGoM coastal flows and in particular the coastal to offshore interactions. As this is a buoyancy dominated area, attention is given to the Mississippi River plume dynamics and the interactions with both the shelf circulation and offshore flows. The numerical simulations employ realistic atmospheric forcing (COAMPS 27km). The effects of wind on the development and evolution of the Mississippi River plume are studied in detail, together with purely wind-driven currents and changes in sea level. The effect of strong topographic features, such as the DeSoto canyon are also addressed. Process oriented simulations in the absence of wind forcing produce coherent shelfbreak flows eastward of the Mississippi River delta, which are able to transport the Mississippi River plume eastward onto the deep areas of the DeSoto Canyon, where the plume waters may interact with mesoscale eddies and be subject to offshore removal. Alongshore coastal flows are observed over the shelf in the presence of eastward/westward winds, which is followed by set-up and set-down of sea surface height and also alongshelf transport of Mississippi River waters.
Successful simulations of circulation in the Northern Gulf of Mexico require improvements in the outer, GoM-HYCOM model. Satellite products are employed to demonstrate improvements in the prediction of the Loop Current extension and the associated eddy field through data assimilation. These products include ocean color data, time series and maps of Sea Surface Temperature (SST), Sea Surface Temperature Residuals (SSTR), Sea Height Anomaly (SHA) and Sea Height Residuals (SHR) and spectrum and wavelet analyses of the SSTR and SHR time series. Model validation includes comparisons of the LC frontal location, based on the depth of the 20°C isotherm, derived by temperature profiles obtained from the GoM-HYCOM model and derived from satellite altimetry observations.
Number 92 - Session 5
ASSIMILATION OF TEMPERATURE IN THE NORTHWEST PACIFIC OCEAN MODEL USING THE ENSEMBLE KALMAN FILTER
Gwang-Ho Seo1, Sangil Kim2, Byoung-Ju Choi3, Yang-Ki Cho1, Young-Ho Kim 4, Chang-Sin Kim1
1Department of Oceanography, Chonnam National University, Gwang-Ju, Korea
2The College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, Oregon, USA
3Department of Oceanography, Kunsan National University, Gunsan, Korea
4Korea Ocean Research & Development Institute, Ansan, Korea
Abstract
The Ensemble Kalman Filter (EnKF) was implemented into a regional ocean modeling system for accurate ocean prediction. The accurate prediction on sea water temperature and currents provides valuable information to weather prediction, fisheries, material transport and reduction of natural disaster. Using the Regional Ocean Model system (ROMS), we developed successfully a unified grid system for marginal seas of the Northwest Pacific Ocean covering four different regions, the northwestern Pacific Ocean, the East China Sea, the Yellow Sea, and the East/Japan Sea. To improve the short-term prediction of ocean circulation in the Northwest Pacific Ocean, data assimilation has been implemented.
After the initial condition was perturbed using multivariate Empirical Orthogonal Function (EOFs), the satellite-observed Sea Surface Temperatures (SST) were assimilated into the numerical ocean model every 7 day from 1 January to 31 December 2003. For the open boundaries the model results from a global ocean model, Estimating the Circulation and Climate of the Ocean (ECCO), was used. The surface forcing was provided from the European Center of Medium Range Weather Forecasting (ECMWF) reanalysis data. The ensemble spread dramatically decreased in a few assimilation steps due to the large false correlation between distant state variables. Localization of error covariance matrix was adapted to resolve the ensemble collapse. Result of data assimilation was verified by comparing model results with observational data such as SST and sea surface height. The assimilation scheme enhances the consistency between the model and the assimilated data and improves the short-term predictability as well. This preliminary work shows promising results that this assimilation scheme can be used in establishing operational ocean prediction systems for the Northwest Pacific marginal seas.
Number 61- Session 4
PERFORMANCE OF FOUR SEA SURFACE TEMPERATURE ASSIMILATION SCHEMES IN THE SOUTH CHINA SEA
Shu Yeqiang1, Zhu Jiang 2, Xiao Xianjun 3, Yan Changxiang2, Wang Dongxiao 1
1, Key Laboratory of Tropical Marine Environmental Dynamics, South China Sea Institute of Oceanology, Chinese Academic of Science, Guangzhou, China
2, ICCES, Institute of Atmospheric physics, Chinese Academic of Science, Beijing, China
3, Laboratory for Climate Studies, China Meteorological Administration, Beijing, China
Abstract
Four existing sea surface temperature (SST) assimilation schemes are evaluated in terms of their performances in assimilating the Advanced Very High Resolution Radiometer pathfinder best SST data in the South China Sea using the Princeton Ocean Model. Schemes 1 and 2 project SST directly to subsurface temperature according to model-based correlations between SST and subsurface temperature. The difference between these two schemes is related to the order of vertical projection and horizontal optimal interpolation (OI). In Scheme 1, the spatially non-uniform SST observations are first projected to subsurface levels, followed by horizontal OI at each level. While in Scheme 2, the remotely sensed SSTs are first optimally interpolated to all grid points at the surface, followed by projecting gridded SSTs to subsurface levels. Scheme 3 assumes that the mixed layer is well mixed and has a uniform temperature vertically. In Scheme 4, the SST is propagated to subsurface levels using a linear relationship of temperature between any two neighboring depths, which is derived by Empirical Orthogonal Function technique. Since baroclinic modes always exist in a real ocean, we allow projections of out-of-phase relationship in Scheme 4. To verify the results of the four schemes, the authors use the Conductivity-Temperature-Depth data from two cruises during the South China Sea Monsoon Experiment in April and July 1998. It was shown that all four schemes could improve the SST field by reducing about 50% of the Root Mean Square Errors (RMSEs). All but Scheme 3 can improve model thermocline structure that is too diffused otherwise, though the RMSEs increase in the thermocline, especially for Scheme 2. Scheme 3 fails in the subsurface by increasing the thermocline depth, especially when there is a cold model bias. Scheme 4 performs well in the mixed layer by correcting the bias to near zero. Due to better correlation of principal components between neighboring levels, the depth to which SST can be successfully projected downward is deeper in Scheme 4 than in the others. Moreover, with mode decomposition in Scheme 4, projecting temperature with an out-of-phase relationship becomes possible; as a result the results can more correctly reflect the baroclinic modes in the ocean.
Number 21- Session4
ENKF WITH GAUSSIAN ANAMORPHOSIS FOR A 3D OCEAN ECOSYSTEM MODEL: INITIAL RESULTS
E. Simon, L. Bertino
Nansen Environmental and Remote Sensing Center, Bergen, Norway
Abstract
We consider the application of the Ensemble Kalman Filter (EnKF) to a coupled ocean ecosystem model (HYCOM-NORWECOM). Such models, especially the ecosystem models, are characterized by strongly non-linear interactions like the bloom phenomenon and present important limitations for the use of data assimilation methods based on linear statistical analysis schemes. Besides the non-linearity of the model, one is confronted with physical/biological limitations, the analysis state having to be consistent with the model, especially with the constraints of positiveness of some variables. Furthermore the non-Gaussian distributions of the biogeochemical variables break an important assumption of the linear analysis, leading to a loss of optimality of the filter.
We present an extension of the EnKF dealing with these limitations by introducing a non-linear change of variables (anamorphosis function) in order to realize the analysis step in a "Gaussian space". We present also the initial results of the application of this non-Gaussian extension of the EnKF to the assimilation of chlorophyll surface concentration data in a North Atlantic configuration of the HYCOM-NORWECOM coupled model.
Number 93 - Session 4
IMPROVING THE SST AND SSS FORECASTS BY A KALMAN FILTER AUGMENTED CONTROL VECTOR TECHNIQUE: A GLOBAL TEST CASE STUDY USING MERCATOR REANALYSIS DATA.
C. Skandrani, J-M. Brankart, J. Verron, P. Brasseur
CNRS/LEGI/MEOM, Grenoble, France
Abstract
At the beginning of the Global Ocean Data Assimilation Experiment (GODAE), it was acknowledged a need for better ocean observations and ocean forecasts. However, the quality of ocean forecasts largely depends on the forcing quality and particularly on the air-sea fluxes. Here, we want to examine how correcting atmospheric forcing parameters can help improving the quality of the air-sea fluxes and thereby the quality of temperature and salinity forecasts.
Bulk parameterization of turbulent air-sea fluxes remains one of the main sources of error in current ocean model, that strongly affects the ability to provide a realistic forecasting of the mixed layer properties. In this study, a sequential data assimilation approach (derived from the SEEK filter) is developed to control the atmospheric forcing in addition to the ocean state. This is done by augmenting the control vector, which contains the ocean state, by the following parameters that must be controlled: latent and sensible heat flux coefficients, cloud coverage, precipitation, air temperature and air humidity.
A Monte Carlo method is used to simulate the model error covariance around the current state of the system. For that purpose, (1) we postulate a probability distribution for parameter errors: Gaussian pdf of known covariance, parameterized using the natural time variability of the parameters for a period of 7 years (1992-1998), (2) a random sampling of these parameters is done from this pdf (sample size=200), (3) we realize a model simulation for each member of the sample, (4) The covariance of the resulting ensemble simulations is the error covariance in the augmented control space. Order reduction by EOFs decomposition is performed and 50 EOFs are selected which explain 90% of total variance.
As an application, the new scheme is applied to the assimilation of SST and SSS observation sampled from the French operational products of MERCATOR re-analysis data (in the ORCA2 low configuration), and that will serve as a reference to evaluate the performance of the scheme. Two different approaches were performed and compared. In the first one, we only correct the forcing parameters included in the estimation space. The state vector is not corrected and the correction is applied by restarting the model from the initial condition of the current assimilation cycle. The results obtained with this free model run are closed to the MERCATOR re-analysis. In the second approach, the state vector of the model is corrected and the new corrected forcing parameters are taken into account in the next forecast. The results show that this lead to a significant improvement in forecasting of SST and SSS.
Number 38 - Session 2
THE 1/12º GLOBAL HYCOM NOWCAST/FORECAST SYSTEM
O.M Smedstad1, J.A. Cummings2, E.J. Metzger3, H. E. Hurlburt3, A.J. Wallcraft3, J.F. Shriver3
1QinetiQ North America, Planning Systems Inc., Stennis Space Center, USA
2Naval Research Laboratory, Monterey, USA
3Naval Research Laboratory, Stennis Space Center, USA
Abstract
The 1/12 degree global HYbrid Coordinate Ocean Model (HYCOM) nowcast/forecast system has been running in near real time since 22 December 2006 and in real time since 16 February 2007. The Navy Coupled Ocean Data Assimilation (NCODA, Cummings, QJRMS, 2005) system is used to assimilate available observations. NCODA is a fully three-dimensional multivariate optimal interpolation (MVOI) system. The NCODA horizontal correlations are multivariate in geopotential and velocity, thereby permitting adjustments (increments) to the mass fields to be correlated with adjustments to the flow fields. The velocity adjustments are in geostrophic balance with the geopotential increments, and the geopotential increments are in hydrostatic agreement with the temperature and salinity increments. Either the Cooper and Haines (JGR, 1996) technique or synthetic T & S profiles, Fox et al., (JAOT, 2002), can be used for downward projection of SSH and SST. NCODA assimilates remotely sensed sea surface height, sea surface temperature and ice concentration. NCODA also assimilates in situ surface and sub-surface observations from several sources, such as ships, buoys, CTD's, BT's and profiling floats. An incremental updating scheme is used to update the HYCOM forecast variables. The increments can be added over a given number of time steps, ranging from 1 (direct insertion) to the number of time steps in one day. Typically a 6 hour updating interval is used. A NCODA analysis is performed every 24 hours. The daily run consists of a 5 day hindcast and a 1 to 5 day forecast depending on available computer resources. The figure shows the sea surface height field from the system on 15 June 2008. The gray areas represent the ice cover. The different components of the system will be described and the latest results including comparisons with independent observations will be shown.
The prediction system will provide boundary conditions for higher resolution coastal models. An accurate representation of the oceanographic fields at the open boundaries of a coastal model is important for a successful coastal ocean prediction system. Results from the global system can be found on the HYCOM Consortium web page http:/www.hycom.org. The data can also be accessed through this web page.
Number 170 - Session 3
Detecting ocean climate signals using new ice-ocean reanalyses
Gregory Smith, Keith Haines, Alastair Gemmell and Dan Bretherton
University of Reading, Environmental Systems Science Centre (ESSC)
Email: gcs@mail.nerc-essc.ac.uk
Abstract
Ocean assimilation is of great importance for both initialising operational oceanography and coupled models for seasonal forecasting, and for reanalysis of the oceans to detect climatically important signals from the diverse historical ocean data that are available. Our focus is on improving the realism of ocean data assimilation schemes and using assimilation to investigate ocean climate signals. We will demonstrate the benefit of combining Temperature and Salinity assimilation methods and of assimilating S(T) as the observable property. This allows us to exploit the larger spatial and temporal decorrelations of this quantity, compared with S(z), allowing flow dependent assimilation and recovery of water mass information.
The S(T) algorithm has been implemented into the NEMO global ice-ocean model and two reanalyses have been made: a 50-year reanalysis at 1° resolution, and a 21-year reanalysis at 1/4° resolution. Overall, the assimilation is able to prevent drifts in many ocean metrics, and brings the model in better agreement with accepted values. An evaluation of water mass properties in various ocean syntheses (e.g. ECCO, MERCATOR, ECMWF, SODA) performed as part of the CLIVAR-GSOP intercomparison, shows that the S(T) reanalyses provide excellent agreement with in situ observations.
Number 34 - Session 5
OPERATIONAL OCEAN DATA ASSIMILATION SYSTEM FOR MONITORING ENSO AT JAPAN METEOROLOGICAL AGENCY
Taizo Soga1, I. Ishikawa1, S. Matsumoto2, Y. Fujii2, M. Kamachi2
1Japan Meteorological Agency, Tokyo, Japan
2Meteorological Research Institute, Tsukuba, Japan
Abstract
Japan Meteorological Agency (JMA) has been monitoring El Niño/Southern Oscillation (ENSO) using ocean data assimilation systems since 1995. The products of these systems have also been used as initial conditions for ENSO forecasting. A new version of the ocean data assimilation system has been developed at JMA/Meteorological Research Institute (MRI) since 2000 and it replaced the old one as the JMA operational system in March 2008.
This system consists of an ocean general circulation model and a data assimilation system. We use MRI Community Ocean Model (MRI.COM) as the ocean model and the Multivariate Ocean Variational Estimation (MOVE) System as the data assimilation system. In the ocean model, the horizontal resolution is 1.0º in longitude x 1.0º in latitude (from 75S to 75N), and 50 vertical levels (24 levels in the upper 200m). Atmospheric data from the Japanese Re-Analysis 25 years (JRA-25) and the JMA Climate Data Assimilation System (JCDAS) are used to drive MRI.COM. These atmospheric data allow us to get consistent and high-quality ocean data. The MOVE System assimilates temperature, salinity and sea surface height. The analysis method of MOVE System is a Three-dimensional Variational (3DVAR) method with coupled temperature-salinity Empirical Orthogonal Function (EOF) modes. And Incremental Analysis Updates (IAU) is adopted as the assimilation method. The representation of the structure and dynamics of subsurface temperature (e.g. thermocline in the equatorial Pacific) and salinity were improved by applying these new schemes and increased resolution of the ocean model.
We are going to introduce this new operational system and show the results from this system, especially improvements in the representation of the structure of subsurface temperature and SST forecast skills of ENSO. Additionally, we are going to glance at the developing system including a new ensemble generation method to provide ocean initial conditions for the ENSO prediction model.
Number 176 Session 5
VARIATIONS IN THE MALVINAS CURRENT VOLUME TRANSPORT
A. Spadone and C. Provost
LOCEAN/IPSL, Paris, France
Abstract
The Malvinas/Falkland current is part of the northern branch of the Antarctic Circumpolar Current (ACC). It enters the Argentine Basin after Drake Passage and flows equatorward following the 1000-1500m isobaths along the Patagonian shelf break. It is the main conduit of Pacific water into the Atlantic (cold route of the thermohaline circulation) and it is the link between the ACC and the subtropical gyre in the South Atlantic.
Two data sets of current meter measurements have been collected in the northern part of the Malvinas current below a Jason-1 altimeter track at 40°S-41°S near its merger with the Brazil current offshore Rio de La Plata estuary: the first one during the World Ocean Circulation Experiment (WOCE) from December 1993 to June 1995 and the second during CLIVAR from December 2001 to February 2003.
With the first data set, it was shown that the TOPEX/Poseidon altimeter could be used to sensitively monitor the flow in the upper 1500m when combined with the statistical information on the vertical structure of the current provided by the current meter measurements. The second data set provide new information on the core of the Malvinas current above a depth of 1000m where a mooring had been previously lost.
The two data sets provide coherent means and statistical parameters on the vertical structure of the flow. A 14-year-long time series of Malvinas current volume transport is derived using satellite altimetry.
Over the 14 years, its spectral content exhibited an evolution with, in particular, the apparition, after year 2000, of energy in a spectral band that included the annual period.
We investigate here whether and to what extent the observed interannual variability has a local or remote cause, examining both the wind stress and altimetry variability in the South East Pacific, in the North of Drake Passage and along the path of the Malvinas current.
Number 182 - Session 5
Shelf Seas nowcasting and forecasting using the NEMO model
Dave Storkey1, Enda O'Dea1, Jason Holt2, Pat Hyder1
1 Met Office, Exeter, UK
2 Proudman Oceanography Laborator (POL), Liverpool, UK
Abstract
Operational models of the North-west European Shelf have been running at the Met Office for several years, using GODAE products (the FOAM system) as boundary conditions. The existing operational system is based on the POLCOMS (1) modelling code, but in the future will be transitioned to use the NEMO (2) model. The NEMO model was developed as a deep ocean model but has recently as part of the MERSEA IP project been modified to enable to be used in shelf seas configurations.
Here we present results of tests of the NEMO code in Shelf Seas configuration. Verification against observations is presented for tides-only, tides plus surge and full baroclinic simulations. In each case the NEMO-based system is compared with the equivalent system based on POLCOMS
(1) Proudman Oceanography Laboratory Coastal Ocean Modelling System.
(2) Nucleus for European Modelling of the Ocean
Number 75 - Session 4
MULTI-SENSOR DATA FUSION THROUGH MULTISCALE MODEL: A METHOD FOR HIGH RESOLUTION ORIENTED GEOSTROPHY
J. Sudre1, H. Yahia2 and V. Garçon1
1 LEGOS CNRS , Toulouse, France
2 INRIA, Centre-Bordeaux Sud Ouest, Toulouse, France
Abstract
The Microcanonical Multifractal Formalism (herein referred to as MMF) is a new paradigm for the next generation complex system modeling that allows the derivation, in a data acquisition of turbulent geophysical fluids, of the most significant statistical contents in the framework of reconstructible systems (1). The MMF has been used for evaluating a first order geostrophic function from combined SST and altimetry data (2,3). In this presentation, we propose new methods for solving the difficult problem of orienting the velocity field derived from the MMF applied on SST data. For that matter, we combine high resolution SST data and surface currents using different strategies that illustrate the high potential of the MMF for merging oceanic data acquired at different spatial resolutions. The main idea is to use high resolution SST for deriving an unoriented velocity field, which then receives a coherent orientation from surface currents at lower spatial resolution. The resulting field is an oriented velocity field at the high resolution of SST data. All solution methods are based on a reconstruction formula that provides an efficient way of merging data at different spatial resolutions, and deriving dynamic information from such reconstructed unoriented field.
We provide examples using high resolution SST from MODIS-Aqua (~4km) and lower resolution surface currents (4) coming from a combination of wind-driven Ekman currents derived from QuickSCAT diffusiometer, and geostrophic currents computed from altimetric Sea Surface Heights.
References
(1) A. Turiel, H. Yahia and C. Perez-Vicente, 2008 Microcanonical Multifractal Formalism: a geometrical approach to multifractal systems. Part I: Singularity Analysis Journal of Physics A 41:015501.
(2) J. Isern-Fontanet, A. Turiel, E. Garcia-Ladona and J. Font, 2007. Microcanonical multifractal formalism: application to the estimation of ocean surface velocities. Journal of Geophysical Research, 112:C05024
(3) A. Turiel, J. Isern-Fontanet, E. Garcia-Ladona and J. Font, 2005. Multifractal Method for the Instantaneous Evaluation of the Stream Function in Geophysical Flows. Physical Review Letters, 95:
10450.
(4) J. Sudre and R. A. Morrow, 2008. Global surface currents: a high-resolution product for
investigating ocean dynamics, Ocean Dynamics, vol. 58, DOI 10.1007/s10236-008-0134-9.
Number 24 - Session 4
NORTH-SOUTH SHIFT OF NORTHERN BOUNDARY OF NORTH PACIFIC INTERMEDIATE WATER ON 26.8 σθ SURFACE
S. Sugimoto, K. Hanawa
Graduate School of Science, Tohoku University, Sendai, Japan
Abstract
We investigate temporal changes of North Pacific Intermediate Water (NPIW) characterized by salinity minimum in the layer from 300 to 700 m depth, utilizing vertically-spatially gridded temperature and salinity datasets produced using Argo float profiles. The NPIW is defined as water mass with salinity S < 34.2, and it is located at the south and east of Japan. The NPIW size variation has an interannual timescale. In order to reveal temporal and spatial behaviors of NPIW, we newly prepare gridded dataset of salinity on the surface of σθ = 26.8 kg m-3 regarded as a core layer of NPIW and thickness of NPIW, by adopting an Optimum Interpolation (OI) method using Argo float profiles. Although the NPIW has almost the same thickness during our analysis period (2003 to 2007), a location of northern boundary of NPIW on the surface of σθ = 26.8 kg m-3 clearly shows the interannual variation. This north-south shift of northern boundary well synchronizes with the NPIW size variation. in the northern boundary region, the oceanic Rossby wave propagation, which might be excited by the wind in the central North Pacific, is observed using AVISO altimetry dataset. It is found that the north-south shift of northern boundary due to the baroclinic Rossby wave propagation is dominant factor in determining the NPIW size variation.
Author's e-mail: sugi@pol.geophys.tohoku.ac.jp
Author's web site: http://www.pol.geophys.tohoku.ac.jp/Exportable/sugimoto/index.html
Number 37 - Session 5
A GLOBAL 4D-VAR DATA ASSIMILATION EXPERIMENT WITH A FULLY COUPLED GCM
Nozomi Sugiura 1, Toshiyuki Awaji1,2, Shuhei Masuda 1, Takashi Mochizuki1,
Takahiro Toyoda1, Hiromichi Igarashi3, Toru Miyama1,Yuji Sasak i1, Yoichi Ishikawa2
1Frontier Research Center for Global Change, JAMSTEC, Yokohama, Japan
2Department of Geophysics, Kyoto University, Kyoto, Japan
3Marine-Earth Data and Information Department, JAMSTEC, Yokohama, Japan
Abstract
A four-dimensional variational (4D-Var) data assimilation system with a fully coupled global ocean-atmosphere-land surface model has been successfully developed with an aim to better define both mean structure and temporal evolution of coupled seasonal to interannual phenomena. Application to state estimation of global climate during the 1990s shows, for example, that the structure in the tropical Pacific region becomes extremely realistic (Fig. 1) and hence the atmospheric condition associated with El Nino events is better represented. An optimal synthesis of observational data and coupled model provides suitable oceanic initial conditions to the El Nino evolution together with the adjustment of coupling intensities regarding the air-sea exchange of fresh water, momentum, and heat. These data well reflect the realistic trend of seasonal to interannual variabilities, thereby enabling us to offer a roughly 1.5 year-long lead predictability for the El Nino. These results suggest that our 4D-Var coupled data assimilation system has an ability to provide superior state representation and forecast potential than earlier assimilation methods.
(Last Updated: 30-10-2008)