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 F - G
Number 132 - Session 3
INCLUSION OF SEA ICE OBSERVATIONS IN AN OCEAN STATE ESTIMATE OF THE LABRADOR SEA USING THE ADJOINT METHOD
I.Fenty, P. Heimbach, C. Wunsch
Massachusetts Institute of Technology, Cambridge, USA
Abstract
Despite being one of the most abundant high-latitude observations of the world ocean, satellite-derived sea ice concentration (SIC) observations are often excluded in global ocean data assimilation efforts. In efforts where the adjoint method, or method of Lagrange multipliers, is utilized, such as in the Estimating the Climate and Circulation of the Ocean (ECCO) project, this exclusion has been due to the difficulty in generating useful adjoints of highly nonlinear sea ice models. Without a sea ice model adjoint, it is impossible to identify the physical pathways linking the sea ice state to the model control variables, such as the atmospheric state and initial ocean state.
Now that a useful sea ice model adjoint has been created, SIC observations can be readily included in state estimation efforts such as ECCO. We demonstrate that additionally including SIC data quantitatively improves a regional one year ocean-sea ice state estimate compared with an estimate made using only observations of in situ ocean temperature and salinity taken from CTDs, profiling floats, and XBTs. The state estimate is made in a regional 1/3 degree coupled sea ice-ocean model of the Labrador Sea using the MIT General Circulation Model.
Number 62 - Session 3
GLORYS - GLobal Ocean ReanalYses and Simulations project:
How lessons learnt from GODAE are used for global ocean reanalyses
N. Ferry1, B. Barnier2,6, S. Belamari3, E. Dombrowsky1, M. Drevillon1, Y. Drillet1, G. Garric1, H. Giordani3, E. Greiner1, L. Parent1, F. Hernandez1, T. Penduff2,6, G. Reverdin4, C.-E. Testut1, A.-M. Tréguier5,6.
1 Mercator-Ocean, Ramonville St Agne, France,
2 MEOM LEGI, Grenoble, France,
3 CNRM Météo France, Toulouse, France,
4 LOCEAN, Paris, France,
5 LPO, Plouzané, France,
6 DRAKKAR consortium.
Abstract
GODAE offered a unique and favourable framework for the development and integration of ocean measurement networks, advanced data assimilation methods and ocean models. It contributed to the sustainability of a global observing system, including satellite and in-situ components. The Jason 2 satellite altimetry mission has been launched, and the global Argo float array now provides thousands of T,S in-situ profiles, completing an invaluable 4-dimensional information of the ocean state. GODAE also supported the improvement of high resolution SST products (GHRSST-PP). The continuous development of advanced data assimilation methods made possible the integration of these multiple observations into ocean general circulation models. Now, operational oceanography centres around the world are routinely producing in real time analyses and forecasts of the state of the ocean at the eddy resolving scale.
If these operational nowcasting/forecasting activities have acquired a remarkable maturity, reanalyses (hindcast experiments) of the "past ocean climate" are still in their early development. Most global ocean reanalyses has been carried out at coarse resolution (~1°) and do not include an accurate estimation of the eddy variability.
We present in this paper the GLobal Ocean ReanalYses and Simulations (GLORYS) project, the aim of which is to construct a set of coordinated simulations, constrained and unconstrained by assimilation of observations, describing at eddy-permitting resolution or better, the space-time evolution of the ocean state and air-sea fluxes in the last few decades. The main objective of the project, which we intend to achieve over the next 3 years (2008-2010), is to carry out global ocean reanalyses over different periods between the 1960s and the present, using the 1/4° resolution ocean/sea-ice global configuration ORCA025, and the present Mercator data assimilation system. ORCA025 is jointly developed by the DRAKKAR consortium and MERCATOR-Océan on the basis of the NEMO code. Our paper will provide details about the actual skills of the model configuration and the assimilation system, and will discuss their new setting appropriate to the long-term scientific objectives of the reanalyses. This paper will also summarise the scientific objectives, the original technical issues, and the working plan of the GLORYS project.
Corresponding authors e-mail: nferry@mercator-ocean.fr, barnier@hmg.inpg.fr
Number 65 - Session 3
OSSE-OSE ACTIVITIES USING THE OCEAN DATA ASSIMILATION AND
PREDICTION SYSTEM, MOVE/MRI.COM
Yosuke Fujii, S. Matsumoto, N. Usui, H. Tsujino, T. Yasuda, M. Kamachi
JMA / Meteorological Research Institute, Tsukuba, Japan
Abstract
MOVE/MRI.COM is the ocean data assimilation and prediction system developed in Japan Meteorological Agency (JMA) / Meteorological Research Institute (MRI). This system is constituted of the ocean general circulation model, MRI.COM, and the data assimilation system, MOVE System. MOVE/MRI.COM has been used in JMA for the operation since March 2008 for monitoring and forecasting of the ENSO and the ocean state around Japan. Here, we show two results of OSSE-OSE activities conducted in MRI using this system.
First, the singular vector analysis is applied to the formation process of the Kuroshio large meander south of
Japan. The purpose of this analysis is to find the effective observation for detecting the sign and predicting the formation accurately. The analysis result shows that an anticyclonic perturbation contacting the Kuroshio path in the southeast of Kyusyu grows rapidly and affects the large meander path two month later. The perturbation produces the cold advection across the Kuroshio in the downstream side and induces the downwelling, resulting in the generation of the anticyclonic eddy in the lower layer. After this eddy is developed, it interacts with the trigger meander: the trigger meander is developed by the baroclinic instability, and it brings about the formation of the large meander path. This result implies that observation data southeast of Kyushu is important for the accurate prediction. Thus, the singular vector analysis allows us to identify an effective position of observations for detecting the target phenomena.
The second activity is an OSE experiment for evaluating the impact of TAO/TRITON arrays on the ENSO forecast. The ocean initial condition for the coupled model, JMA/MRI-CGCM, is prepared by the global version of MOVE/MRI.COM in the JMA ENSO forecasting system. In the OSE experiment, we prepare two data assimilation runs with and without TAO/TRITON data. Although the ocean fields in the two assimilation run is very similar, there are some differences in the zonal gradient of the thermocline in the equatorial Pacific, and the barrier layer thickness, etc. Then, we perform 13-month 11-member ensemble forecast from these two assimilation runs using the JMA's operational coupled model, JMA/MRI CGCM. The Anomaly Correlation Coefficient (ACC) and Root Mean Square Error (RMSE) scores are better in the forecast started from the data assimilation runs with TAO/TRITON data for NINO1+2, NINO3, NINO4 and NINO-WEST (0-15ºN, 130-150ºE) SST indices in the first six months. In the presentation, we additionally
show the result of an OSE experiment for evaluating the impact of ARGO floats data on the ENSO forecast.
Number 55 - Session 3
A KALMAN FILTER AND 3DVAR INTER-COMPARISON WITH NCEP'S NEW OPERATIONAL FORECASTING SYSTEM
Ichiro Fukumori1, David Behringer2, Ou Wang1, Jiande Wang2
1Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, U.S.A.
2National Centers for Environmental Prediction, NOAA, Camp Springs, MD, U.S.A.
Abstract
A Kalman filter is developed for the Modular Ocean Model (MOM4) employed in the latest operational ocean analyses of the National Centers for Environmental Prediction (NCEP), National Oceanic and Atmospheric Administration. The filter employs explicit estimates of model errors and is amenable to assimilating diverse observations that model equivalents can be defined. This study compares the ocean state estimates of this Kalman filter with those of NCEP's 3dVAR method to assess the relative impact of the different approaches on analyzing and forecasting seasonal-to-interannual climate variability.
The Kalman filter is based on that devised by the Consortium for Estimating the Circulation and Climate of the Ocean (ECCO) for near real-time ocean analyses. The ECCO filter employs partitioned, reduced-state, and time-asymptotic approximations of the model state error covariance matrix associated with inaccuracies in atmospheric wind forcing. The filter assimilates temporal anomalies of satellite sea level and in situ temperature profile measurements over the world's oceans. New advancements are devised and implemented including use of adjoint codes in filter operations, improved spatial interpolation around land masks, and robust estimates of vertical heaving motion.
The next operational model of NCEP will be based on MOM4 and will employ a tripolar grid that spans the entire globe including the Arctic Ocean. This model, which is currently undergoing testing, has a nominal 1° grid with a telescoping meridional resolution of 1/3° in the tropics within 20° of the Equator. A 3dVAR method, ported from the current operational MOM3 model, is employed to assimilate temperature and synthetic salinity profiles. The salinity profiles are constructed from the temperature profiles and a local TS relationship. The model error variances are assumed to be proportional to the local temperature and salinity gradients and are computed from the most recent 5-day model average.
This study will present a comprehensive analysis of the filter and 3dVAR estimates implemented with the NCEP ocean model. Utilizing identical models, the analysis provides a unique assessment of the two assimilation methods independent of potential differences in models that are used. The two methods will be briefly described and ocean state estimates will be compared with respect to observations both utilized in the assimilation as well as those independent of the estimates including sea level, temperature and salinity, and ocean mass variations. Differences between the two estimates and a model simulation without data constraints will be examined to assess the relative impact and differences between the two assimilation approaches. Indices of circulation will be analyzed such as strengths of subtropical cells in the Pacific Ocean, meridional overturning circulation of the Atlantic Ocean and the globe, and integrated upper ocean heat content changes of the world's oceans. The assimilations' impact on subsequent variability of the ocean will be assessed by examining model evolution resulting from data assimilated model initialization.
Number 27 - Session 3
PROGRESS TOWARD A 100-YEAR OCEAN REANALYSIS
B. S. Giese1 and J. A. Carton2
1Texas A&M University, College Station, TX, USA
2University of Maryland, College Park, MD, USA
Abstract
Many interesting climate phenomena such as the warming of the North Atlantic and the Dust Bowl Drought occurred or extended into the first half of the 20th Century when the first efforts were made to deploy an ocean observing system. Recently, meteorologists have begun an effort to extend atmospheric reanalyses back to 1900. Motivated by their work, this is a study of how much further back into the historical record it may be possible to extract useful information about ocean circulation using data assimilation. The tool we use for this study is the Simple Ocean Data Assimilation (SODA). We present the results of an effort to extend the SODA reanalysis back to the beginning of the 20th Century. Preliminary results with a simulation show an interesting set of climate variability in the early 1900's, including the resolution of the strong El Nino of 1918. In another set of experiments we examine the impact of the reduced quality of surface forcing information by adding noise to the winds, heating, and freshwater flux. The analysis focuses on determining what useful information can be extracted from the ocean observations, and on what space and timescales.
Number 173 - Session 2
MODEL SALINITY PRODUCT CHARACTERIZATION FOR THE SMOS SATELLITE MISSION
J. Gourrion1, A. Aretxabaleta1, J. Ballabrera-Poy1, B. Mourre1, J. Font1
1Institut de Ciències del Mar -CSIC, SMOS Barcelona
Expert Centre on Radiometric Calibration and Ocean Salinity, Barcelona, Spain
Abstract
Availability of operational GODAE products during recent years can provide improved salinity variability characterization. A comprehensive understanding of the salinity variability is fundamental for the development and validation of satellite salinity observations for the SMOS (Soil Moisture and Ocean Salinity) mission. Sea-surface salinity (SSS) is extracted from global 1/12 degree resolution HYCOM archived solutions (Nov03-Jun08). When compared with ARGO profiler data the model salinity exhibits a small global bias (0.02 psu) that increases substantially when estimated for individual basins (e.g., North Atlantic, -0.19 psu).
Also, a requirement for the generation of SMOS Level 3 global SSS products through optimal interpolation is the availability of SSS spatial correlations. These correlations are calculated from the global model solutions resulting in correlation scales of up to 2000 km in the equatorial Pacific. The model salinity annual cycle exhibits greater magnitudes associated with large rivers and in the subtropical gyres. The smallest annual cycle is found in the Antarctic Circumpolar Current. While the temperature annual cycle phase lags exhibit the characteristic winter-summer dipole, the salinity phase lags display a more complex latitudinal and longitudinal structure.
Number 88 - Session 5
THE BLUELINK ANALYSIS AND RE-ANALYSIS SYSTEMS AT WORK IN AUSTRALIA
David A Griffin, Madeleine L. Cahill, Peter R Oke, Ken Ridgway, Andreas Schiller
Centre for Australian Weather and Climate Research: A partnership between CSIRO and the Bureau of Meteorology
Abstract
The Australian Bluelink project is a three-way partnership between the CSIRO, The Bureau of Meteorology and the Royal Australian Navy. As a result of this project, Australia now has a short-term, eddy-resolving ocean forecasting system run operationally at the Bureau of Meteorology (OceanMAPS), a rapidly-deployable nested forecasting system implemented on Navy computers (ROAM), an eddy-resolving ocean Reanalysis (BRAN) spanning the altimeter era, and a suite of model-independent reanalysis and analysis products (HRRA). The HRRA and BRAN have existed longer (since 2004, and 2006, respectively) than either ROAM (2008) or OceanMAPS (2007), so they have been the first to be applied to problems requiring accurate and detailed estimates of ocean currents.
The High Resolution Regional Analysis (HRRA) is based on a simple two-dimensional mapping of sea level measured by altimeters (Topex/Poseidon, Jason-1, ERS1, ERS2, Envisat and GFO) and coastal tide gauges. Daily-updated maps of sea level, geostrophic currents computed from the gradients of sea level, and sea surface temperature have been freely available at http://www.marine.csiro.au/remotesensing/oceancurrents/ since 2004. These are used by a wide range of sporting (yachtsmen, fishermen, open-ocean rowers), engineering (oil exploration and extraction, shipping), public sector (defense, search-and-rescue, law enforcement, fisheries management) and scientific users.
The Bluelink ReANalysis (BRAN) is the result of a much more complex system (see http://www.cmar.csiro.au/staff/oke/BRAN.htm). Here, a global ocean circulation model that is eddy-resolving (10km resolution) in the Australasian region is constrained to be as faithful to reality as possible by assimilating all the available observations from satellites (sea level, sea surface temperature) and in-situ platforms (Argo floats measuring temperature and salinity to 2000m, moorings, etc). BRAN has been made freely available (but only for research purposes) in order to encourage evaluation and application to a wide range of questions. Examples include assessing the interconnectedness of fisheries management zones due to ocean transport of larvae, back-tracking the paths taken by derelict fishing nets that accumulate on Australia's northern shores, and mapping Australia's Ocean Renewable Energy resources. The system was even used to help solve a 67-year-old mystery: the resting place of the pride of Australia's WWII naval fleet, HMAS Sydney.
Number 151 - Session 2
THE EUMETSAT OCEAN AN SEA ICE SAF (OSI SAF) :
A CONTRIBUTION TO OPERATIONAL OCEANOGRAPHY
Guenole Guevel
Météo-France, Centre de Météorologie Spatiale, Lannion, Brittany, France
Abstract
The EUMETSAT OSI SAF (www.osi-saf.org) was created in 1997 as an answer to requirements from the meteorological and oceanographic communities of EUMETSAT Member States and Co-operating States for a comprehensive information derived from meteorological satellites at the ocean-atmosphere interface.
The OSI SAF consortium is constituted of Météo-France as leading entity, and Met.no (Norske Meteorologiske Institutt), DMI (Danish Meteorological Institute), SMHI (Swedish Meteorological and Hydrological Institute), KNMI (Koninklijk Nederlands Meteorologisch Instituut) and IFREMER (Institut Français de Recherche pour l'Exploitation de la MER) as co-operating entities.
The two previous phases, the Development phase (1997-2002) and the IOP (initial Operations Phase, 2002-2007) met the main target which was to develop, validate and then produce operationally quality controlled satellite-derived products related to four key parameters (Sea Surface Temperature, Radiative Fluxes, Sea Ice, Wind) over various geographical coverage from regional to global.
These products are currently available in near real time both through EUMETCAST and local FTP servers and off line from local archive. The archiving at EUMETSAT UMARF is being implemented.
The current phase of the OSI SAF, the CDOP (Continuous Development and Operations Phase) has taken into account new requirement sources, in particular from GODAE, GHRSST and GCOS at international level, and GMES (through MyOcean) at European level, with a strong need for increasing the temporal and geographical resolution of the products and for extending the coverage range from coastal to global. Its main aspects are :
- extension of the global products range : The OSI SAF already produces products at global coverage : the MetOp SST, the Sea Ice, and the Winds from MetOp/ASCAT and QuikSCAT/SeaWinds
- new products ( Sea Ice emissivity, Sea Ice Drift),
- temporal and geographical resolution increasing, with access to some of the products at full resolution, through flexible extraction interface tool on FTP servers, such as NAIAD (in complement with the existing EUMETCAST dissemination of products over predefined areas and projections),
- the preparation for the use of new satellites ( NPP-NPOESS, MTG, GOES-G, post EPS) and new optional parameters, such as Ocean Colour,
- use of new satellites such as SENTINEL-3 under consideration
The objective of this paper is to offer an overview on the OSI SAF project, The Users requirements, the target production at the end of the CDOP, the (pre-)operational production status, and the preparation for future satellites.
Number 153 - Session 4
ON THE USE OF SATELLITE ALTIMETER DATA IN ARGO QUALITY CONTROL
S. Guinehut1, C. Coatanoan2, A.-L. Dhomps1, P.-Y. Le Traon2 and G. Larnicol1
1CLS, Space Oceanography Division, Ramonville, France
2Ifremer, Technopole de Brest-Iroise, Plouzane, France
Abstract
A new method has been developed to check the quality of each Argo profiling floats time series. It compares collocated Sea Level Anomalies (SLA) from altimeter measurements and Dynamic Height Anomalies (DHA) calculated from the Argo temperature (T) and salinity (S) profiles. By exploiting the correlation that exists between the two data sets along with mean representative statistical differences between the two, the altimeter measurements are used to extract random or systematic errors in the Argo float time series. Different kinds of anomalies (sensor drift, bias, spikes, etc) have been identified on some real-time but also delayed-mode Argo floats.
An example of SLA/DHA time series is given on Figure 1 for float number 1900249. It shows clearly a positive drift of the DHA time series regarding the SLA time series as the float is traveling from East to West in the Tropical Atlantic Ocean. It seems that one of the float sensors presents a malfunction and it has to be carefully checked by the PI.
Figure 1: SLA and DHA time series for float number 1900249 (in cm). The geographical position of the float is also indicated, the bleu cross corresponding to the deployment position and the red cross to its last reported position.
(Last Updated: 30-10-2008)