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 T - W
Number 135 - Session 4
REGIONAL IN-SITU OBSERVATORY: GLIDER OPERATIONS IN THE NORTHWESTERN MEDITERRANEAN SEA IN WINTER 2008 (EGO)
P. Testor1 , L. Mortier1 , B. Lheveder1, V. Taillandier1, U. Send2, R. Davis2, D.Smeed3,L.Merckelbach3,, A. Alvarez4, J. Tintore4, B. Casas4, S. Ruiz4, P. Lherminier5, T. Terre5, G. Krahmann6, J. Karstensen6 , F. Bourrin7, H. Claustre7, J.-J. Naudin8, V. Rigaud9, T. Carval10, L. Petit de la Villeon10, C. Jones11, J. Sherman2
1 LOCEANIPSL, Paris, France
2 SIO, La Jolla, CA, USA
3 NOCS, Southampton, UK
4 IMEDEA, Esporles, Spain
5 LPOIFREMER, Brest, France
6 IFM/GEOMAR, Kiel, Germany
7 LOV, Villefranche, France
8 OOB, Banyuls/mer, France
9 SSM-IFREMER, La Seyne/mer, France
10 Coriolis-IFREMER, Brest, France
11 Webb Research Corp., Falmouth, MA, USA
Abstract
It has been envisaged that individual or even fleets of gliders that are equivalent to remotely steered profiling floats deployed at both regional and large scale could efficiently complement the ARGO array for the benefit of both operational oceanography and academic research including in particular, "green ocean" objectives. The EGO group, gathering together several teams of engineers and scientists, tries to demonstrate this vision through pilot projects. The objectives of the EGO group are to share the corresponding efforts and to provide support to people operating gliders. EGO also helps to coordinate deployments of enough gliders for dedicated exercises in targeted areas.
As a first EGO demonstration and in the framework the EU project MERSEA, as well as national projects, seven deep (1000m) and two shallow (200m) gliders were used between January and June 2007 to collect temperature and salinity profiles as well as oxygen, backscattering, Chlorophylla, and CDOM fluorescence on the gliders equipped with biogeochemical sensors. One deep glider was devoted to the large scale monitoring of the gyre in its longitudinal extension from the Balearic Islands to the Gulf of Genova while another has crossed the basin in the meridional direction. A shallow glider was deployed to cross the channels between the Balearic Islands and a deep one sampled the Ligurian Sea offshore of Villefranche s/Mer to monitor an upstream section on the Northern Current. Three deep gliders were devoted to the investigation of the deep convection area in the Gulf of Lion. Finally, a shallow glider was deployed to study mesoscale and submesoscale features associated with the Northern Current. In winter 2008, a similar experiment (EGO2008) has been carried out with more gliders. From late january 2008 to the end of April 2008, up to 9 gliders (7 deep Slocum and 2 shallow Slocum) were operating simulataneoulsy at sea.
The gliders were deployed and recovered from small boats from the "gliderports" hosted in marine stations around the basin. Some of the gliders that were equipped with alkaline batteries were recovered to change their batteries and were immediately redeployed. Remote steering of the gliders and real time data download were allowed by Iridium bidirectional link and carried out using "groundstation" computers based in the partners institutions. Routines have been developed for real time data processing and visualization on the web in order to better control the glider missions.
Moreover, forecasts of the trajectories of the fleet based on the currents produced by MERCATOR on a weekly basis, were published on the web both to help the pilots reshaping the network periodically and also to deliver the information on the glider positions to the Spanish and French marine security agencies.
A technological and scientific assessment is now being undertaken. A first obvious conclusion is that operating several gliders over long periods, requires multisite coordinated facilities: operators are needed to 1) evaluate the actual trajectories of the gliders and modify them if necessary, 2) modify some parameters (speed, sampling rate, ....) or correct small malfunctions of the gliders, and 3) quickly organize an emergency recovery in case of a major failure. These first EGO coordinated deployments of glider fleets were a successful technological demonstration and the amount of data collected demonstrates how intensively one can observe the ocean interior using such platforms in a particular area.
Number 105 - Session 4
VERTICAL STRUCTURES OF THE NORTH PACIFIC MODE WATERS
K. Toyama1, and T. Suga1, 2
1Tohoku University, Sendai, Japan
2IORGC/JAMSTEC, Yokosuka, Japan
Abstract
Vertical structures of mode waters in the North Pacific are investigated using the Argo data from 2001 until 2007. As parameters of vertical structures, vertical gradient of water properties and Turner angle are used. Four mode waters in the North Pacific, Subtropica l Mode Water (STMW), Eastern STMW (ESTMW), Central Mode Water (CMW) and Transition Region Mode Water (TRMW) are defined as thick (> 100 dbar) low potential vorticity (PV < 2.0 x 10-10 m-1 s-1) waters with specific water properties within specific areas.
Temperature gradient and density gradient of SMTW is in linear relationship and salinity gradient of STMW is very small. Density gradient of STMW is mostly determined by temperature gradient. On the other hand, gradient of salinity and that of temperature of CMW, ESTMW and TRMW are relatively large and take various values in wide ranges. It is suggested that these three mode waters have density compensating stratification of temperature and salinity.
In the low-PV portion of STMW, Turner angle is less than 60 degree and vertically homogeneous. Tu in the low-PV portion of CMW is also mostly homogeneous with 60 ? 70 degree. Tu of ESTMW and TRMW, on the other hand, have different structure in upper and lower side of low-PV portion. In the lower side of ESTMW and in the upper side of TRMW, Tu is larger than 70 degree on average, indicating active salt finger type convection. In the upper side of ESTMW and in the lower of TRMW, Tu is less than 70 degree, suggesting relatively stable stratification.
Salt finger type convection possibly modifies temperature and salinity of ESTMW and TRMW from the lower and upper side of those, respectively. T/S modification by double diffusive convection appears relatively small in case of STMW and CMW.
Turner angle at the cores of pycnostads on the θ-S plane. Circle denotes positive Tu and triangle denotes negative Tu
Number 114 - Session 4
THE GLOBAL 1/12° MERCATOR OCEAN FORECASTING SYSTEM: SCIENTIFIC DESIGN AND FIRST RESULTS
Benoît Tranchant1, C-E. Testut2, R. Bourdallé-Badie1, C. Derval1, O. Le Galloudec3 and Y. Drillet3
1CERFACS,Toulouse, France
2MGC incorporation, Toulouse, France
3Mercator Ocean, Toulouse, France
Abstract
An integrated ocean forecasting and data assimilation system has been and is continuing to be developed by Mercator Ocean. It consists of a global ocean and sea ice high resolution model with a horizontal resolution of 1/12° and 50 vertical levels based on the NEMO OGCM which is coupled to the data assimilation scheme named SAM2v1 (based on the SEEK filter). Assimilation of both in situ and remotely sensed data (SLA, SST) provides the initial conditions required for numerical ocean prediction.
A first step has been to perform an interannual simulation forced by ECMWF atmospheric forcing without data assimilation. This step was essential to calculate statistics of prognostic variables used to model the background error covariances into the data assimilation scheme.
This paper summarizes various considerations taken into the design of this ocean forecasting system, and describes its main characteristics. A special focus will be given on (i) the interannual simulation, (ii) the data assimilation parameters and (iii) the first hindcast experiments.
Number 152 - Session 5
ASSESSING THE CONTRIBUTIONS TO RECENT SEA LEVEL CHANGE BY COMBINING ALTIMETRIC DATA AND A GLOBAL OCEAN MODEL
F. Trumm1, M. Wenzel1, J. Schröter1
1Alfred Wegener Institute, Bremerhaven, Germany
Abstract
Using 4DVar data assimilation, a variety of datasets is used to constrain a global ocean model. The aim is to differentiate between the different contributions to recent (1993 to 2003) sea level changes. TOPEX/Poseidon altimetric data constrains the free surface of the ocean model and hence the total sea level changes. A high resolution temperature dataset down to a depth of 700 m (Willis et al, 2004) is assimilated. Given that sea level height and thermosteric sea level changes are constrained, this leaves us with a good estimate for the mass input as the residual of the two. We present results from the optimized solution showing the global mean as well as regional structures of sea level change. In a more detailed analysis we compare model output to the observed sea level changes. Finally, a comparison of 2002/2003 model results with GRACE data is used to assess the quality of our solution.
Number 67 - Session 3
REPRODUCIBILITY OF CURRENTS AND WATER MASSES IN THE NORTH PACIFIC SUBARCTIC REGION IN MOVE/MRI.COM
Norihisa Usui1, S. Ishizaki2, Y. Fujii1, H. Tsujino1, M. Kamachi1
1JMA / Meteorological Research Institute, Tsukuba, Japan
2Japan Meteorological Agency, Tokyo, Japan
Abstract
The ocean data assimilation and prediction system, MOVE/MRI.COM, has been developed in Japan Meteorological Agency (JMA) / Meteorological Research Institute, and has been used for the operation in JMA since March 2008. The system is composed of the ocean general circulation model (MRI.COM) and the data assimilation system with a variational analysis scheme (MOVE). We evaluate reproducibility of assimilated results of MOVE/MRI.COM, especially focusing on the North Pacific subarctic region.
Southern boundary of the western North Pacific subarctic region, so-called the Kuroshio-Oyashio confluence zone, exhibits energetic mesoscale eddy activity, which makes difficult to give a realistic reproduction of the ocean state in assimilation. Distributions of the Oyashio water and mesoscale eddies in the assimilated field are reproduced well. We however find some issues in reproducing the Oyashio water. In the assimilated field, the Oyashio water shows too low temperature (lower than 0°C) in some situations. In addition, the assimilated field tends to overestimate the area of the Oyashio water in summer. These issues arise from a non-Gaussian distribution of temperature of the Oyashio water. Improvements of the analysis scheme to solve these issues and their impacts will be presented.
In the North Pacific subarctic region, the mesotherrmal structure, the shallow temperature minimum (dichothermal water) and underlying temperature inversion (mesothermal water), is the characteristic structure. Using reanalysis dataset of the North Pacific version of MOVE/MRI.COM from 1985 to 2006, we evaluate the mesothermal structure in the assimilated field. The mean climatological state of the reanalysis dataset gives a realistic distribution and seasonal cycle of these water masses. Interannual valiability of these water masses will also be presented.
Number 166 - Session 4
OPERATIONAL ASSESSMENT OF THE REAL-TIME MERCATOR OCEAN ANALYSES AND FORECASTS
Nathalie Verbrugge2, Nicolas Pene2, Eric Greiner2, Marie Drévillon1, Jean-Michel Lellouche1, Mounir Benkiran2, Lucas Nouel2, Gaëtan Vinay3
1CERFACS, Toulouse, France
2CLS, Toulouse, France
3MGC, Toulouse, France
Abstract
Since the beginning of GODAE and also in the framework of the European projects MERSEA and now GMES/MyOcean, Mercator-Ocean has been designing a hierarchy of ocean analysis and forecasting systems based on numerical models of the ocean and data assimilation systems. Since April 2008, Mercator runs a global ocean configuration at ¼° horizontal resolution and a North Atlantic and Mediterranean zoom at 1/12°, both having 50 levels on the vertical with a surface refinement. The real time operation of these systems produces each week realistic 3-dimensional oceanic conditions (temperature, salinity, currents...) two weeks including a hindcast and forecast.
Moreover, the regional zoom is operated daily to produce 7 days forecasts with daily updates of the atmospheric ECMWF forcing. Before the distribution of its real-time products, Mercator-Ocean checks the quality with a calibration/validation procedure. This assessment procedure is operated automatically. Outputs are compared to independent observations (Cersat Sea Ice concentration and drifts, Odyssea High resolution SST, Coriolis drifters, T/S climatology, historical data sets ...). Hindcast, nowcast and forecasts are inter-compared to check the consistency of the production. Some MERSEA metrics (mooring points and 2D oceanic sections) are also used.
The assimilation performance is monitored in the observations space (in situ Coriolis profiles, ½° SST RTG, and altimeters) and the delivery quality of the input data sets is supervised. To fulfil this operational assessment of the system outputs, more than 1300 graphics per week are produced and visualised by an operator team through web pages. A step-by-step document guides them through the validation process.
Corresponding authors e-mail: nverbrugge@mercator-ocean.fr
Number 184 - Session 3
Global hydrographic variability patterns during 2003-2007
K. von Schuckmann, F. Gaillard and P.-Y. Le Traon
IFREMER, Brest, France
Abstract
Global temperature and salinity fields from the near surface layer down to 2000m depth based on ARGO measurements are used to analyze large-scale variability patterns on annual to interannual time scales as they are derived from the monthly mean values during the years 2003- 2007. Previous estimates of global hydrographic fluctuations have been created using different data sets, partly based on scarce sampling. The substantial advantage of this study includes a detailed summary of annual to interannual variability patterns of the global ocean based on a single and more uniform data base. The dominant signal of upper ocean variability is the annual cycle characterized by a clear hemispheric asymmetry with strongest amplitudes in the northern hemisphere where changes of the annual signal from year to year are strong. Annual amplitudes of the salinity field play an important role in the near surface tropical and subpolar parts of the global ocean. The dominant harmonic of the global temperature field increases at mid-latitudes from the surface down to more than 300m depth and amplitudes are subsurface intensified in the tropical basin. At mid-latitudes in the area of the subtropical and subpolar fronts, interannual anomalies show a deep baroclinic component down to more than 1500m depth in all three ocean basins. In the tropical basins, dominant patterns of baroclinic variability are mostly confined to the upper 500m depth. However, the dominant signatures of global interannual variability occur in the equatorial band and between 5-10±N during the years 2003-2007. A global estimation of long-term changes of heat and salt content as well as of steric height are derived from the gridded field. Heat content and steric height changes are clearly associated with a positive trend during the 5 years of measurements. Salinity content is dominated by interannual fluctuations. Steric height deduced from in-situ measurements are in reasonable agreement. Although some discrepancies remain they are smaller than in previous results.
Number 120 - Session 4
Influence of wind stress error in the 4DVAR basin scale ocean circulation analysis
T. Wakamatsu
Fisheries and Oceans, Canada
Abstract
The effects of the parameterized wind stress error covariance function on the a priori error covariance of an ocean general circulation model (OGCM) are examined. These effects are diagnosed by computing the projection of the a priori model state error covariance matrix to sea surface height (SSH). The sensitivities of the a priori error covariance to the wind stress curl error are inferred from the a priori SSH error covariance. They are shown to differ between the subpolar and subtropical gyres due to different contributions from barotropic and baroclinic ocean dynamics. The spatial structure of the SSH error covariance due to the wind stress error indicates that the a priori model state error is determined indirectly by the wind stress curl error. The impact of this sensitivity on the solution of a four dimensional inverse problem is inferred.
Number 59 - Session 5
Seasonal ensemble EnOI experiment in Pacific by using HYCOM
L.Y. Wan1, J. Zhu2 and L. Bertino3
1National Marine Environment Forecasting Center , Beijing, China
2 Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
3 Mohn-Sverdrup Center,Nansen Environmental and Remote Sensing Center, Bergen, Norway
Abstract
A simplified version of Ensemble Kalman Filter, the so-called Ensemble Optimal Interpolation (EnOI) requires only a small fraction of the computing cost of the EnKF. The EnOI analysis is computed in the space spanned by a stationary ensemble of model states sampled during a long-time integration. Overlarge ensemble spread usually overestimates the instantaneous variability. Therefore, a parameter is used by reduced them.
Seasonal ensemble without seasonal variability has been used in EnOI experiment in Pacific. The experimental parameter is erased and the seasonal characters have been preserved. The results are validated against merged satellite and in situ data Global Daily Sea Surface Temperatures in the global ocean (MGDSST) and the independent Argo temperature and salinity profiles and show that the seasonal ensemble EnOI method performs better than the experimental parameter EnOI.
Number 49 - Session 5
DESIGN ASSESSMENT OF AN ON GOING IN-SITU OCEAN OBSERVING SYSTEM IN CHINESE MARGINAL SEAS
Ruiwen Wang1, Jiang Zhu2, Yiguo Xiao1, Xiaobao You1
1ICCES of Institute of atmospheric physics, Chinese academy of sciences, Beijing, China
2LAPC, of Institute of atmospheric physics, Chinese academy of sciences, Beijing, China
Abstract
Chinese Academy of Sciences is now developing a sustainable ocean observing system in the Chinese marginal seas (including Yellow Sea, East China Sea and South China Sea). This observing system includes three coastal stations, four offshore buoys and regular cruise sections that will be carried out by its research vessels. The locations of these buoys and cruises are now determined. Implementations of this observing system will be carried out in the following years (2008-2011). In this paper a design assessment of this observing system in the context of monitoring large scale interseasonal and interannual variability is discussed using an existing Kalman theory based method of Sakov and Oke (2007). This approach examines the error reductions made by assimilation of observations from the given buoy and cruise sections according to the Kalman filter theory. The background error covariance is calculated by ensemble of model based results. Two sets of ensemble are used from which the stationary background error covariance matrices can be calculated. One is from an assimilation run of a coarse resolution (0.5° by 0.5°) ocean model. Another is from a higher resolution (0.12° by 0.12°) model simulation. The results from the two datasets are similar. We found that the SST observations from the three coastal stations and four offshore buoys can effectively reduce the SST analysis error in Bohai Sea, Yellow Sea, East China Sea and northern and middle parts of South China Sea. However, in the regions east of Viet Nam and east of Taiwan and Philippine, the percentages of error reductions are relative small.
Number 73 - Session 2
OPERATIONAL ISSUES WITH BLUElink> OCEAN MODEL, ANALYSIS AND PREDICTION SYSTEM (OceanMAPS)
Graham Warren1, Gary Brassington1, 2, Mikhail Entel1, Xinmei Huang1, Leon Majewski1, Tim Pugh1, 2, Claire Spillman1, 2
1 Bureau of Meteorology, Melbourne, Australia
2 Centre for Australian Weather and Climate Research, Melbourne, Australia
Abstract
The BLUElink> Ocean Model, Analysis and Prediction System (OceanMAPS) has been operational at the Bureau of Meteorology since August 2007. This poster will highlight some of the issues that have been encountered in establishing a routine, robust ocean forecast system and the delivery of products.
We will describe systems which collect, monitor, quality control both in situ and remotely sensed data which is used in the assimilation phase of the system. In situ data are sourced from the GTS and the GODAE data servers (USGODAE and Coriolis) and include temperature and salinity data from Argo profiling floats, Ship of Opportunity lines and moored buoys. The key satellite data sets are altimetry from JASON 1 and ENVISAT and sea surface temperature from AMSR-E. Procedures are being developed for the receipt of JASON 2 data.
The scheduling of the OceanMAPS system has been integrated into the routine operations at the National Meteorological and Oceanographic Centre (NMOC) which runs the Bureau of Meteorology's Numerical Weather Prediction models and other numerical forecasting related tasks. The timing of OceanMAPS and impacts and dependencies with other systems will be discussed.
Product servicing is aimed at a number of different levels of user. An on-line graphical service on the Bureau's public web site allows general access to charts of daily averages of sea surface temperature, sea surface anomaly and sea surface salinity. Three dimensional grids of the latest output can be obtained by subscription from the ftp server. Products are being made available to the research community through an OpenDAP server and there are plans to extend this capability to more general users of the data.
Number 71 - Session 5
OPERATIONAL FORCASTING SYSTEM OF THE JAPAN SEA FOR FISHERY ENVIRONMENTS
T. Watanabe1, K. Takayama2, D. Simizu1 and N. Hirose2
1 Japan Sea National Fisheries Research Institute, Niigatqa, Japan
2 Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan
Abstract
Distributions of fisheries resources are closely related to the ocean conditions in the surface layer (e.g. water temperature, current velocity, allocation of fronts and warm/cold eddies). Hence, the in-situ CTD observation network along the Japanese coastal area in the Japan Sea is maintained by the national and the prefectural fisheries research institutes of Japan. However, in-situ CTD data observed by research vessels are highly limited in space and time, especially offshore area. In this study, a numerical forecasting system with the data assimilation of the sea surface temperature (SST), the sea surface height anomaly (SSH) and CTD is introduced for reproducing and forecasting the oceanographic conditions of the Japan Sea continuously in space and time.
The Japan Sea forecasting system with the data assimilation of both SST and SSH has been developed by RIAM, Kyusyu University (Hirose et al., 2007); an approximate Kalman filter method for SSH and a simple nudging method for SST were applied to a three-dimension general circulation model (RIAM ocean model). The horizontal and vertical resolutions of the model are 1/12 degree and 36 layers, respectively. In this study, dynamic height calculated from each CTD profile is assimilated into the model using the approximate Kalman filter (Takayama et al., 2008). In-situ CTD data are summarized by the Japan Sea National Fisheries Research Institute in quasi real time and the reproducing and the forecasting within 60 days are calculated once a week operationally.
As a result of adding the CTD assimilation scheme, the accuracy of reproducibility increases. For example, the RMS difference of the temperature at the depth of 100m between the observation and the numerical result with the assimilation of only SST and SSHA is estimated as 1.00 degree. On the other hand, that with the assimilation of SST, SSH and in-situ CTD is estimated as 0.95 degree. In particular, because SSH is inaccurate in the coastal area, in-situ CTD is highly effective in reproducing coastal ocean conditions (see Figure 1). This improvement is also useful for coastal fisheries. The result of this operational forecasting system of the Japan Sea is now open in the web site (http://jade.dc.affrc.go.jp/), which is called as the JADE (JApan sea Data assimilation Experiment) system.
Figure 1. Temperature distribution at the depth of 200m of (a) the assimilation using SST and SSH, (b) the assimilation using SST, SSH and CTD.
Number 15 - Session 4
AN ENSEMBLE THREE-DIMENSIONAL VARIATIONAL DATA ASSIMILATION SYSTEM FOR THE GLOBAL OCEAN
N. Daget1, A. T. Weaver2 and M. A. Balmaseda3
1CERFACS, Toulouse, France
2CERFACS, Toulouse, France
3ECMWF, Reading, United Kingdom
Abstract
A three-dimensional variational data assimilation (3D-Var) system has been developed for global analysis with the OPA ocean general circulation model. The global 3D-Var system is based on an earlier system developed for the tropical Pacific, but has been extended to incorporate new features including fully multivariate background-error covariances and the capacity to produce ensembles of ocean analyses for climate studies and forecast initialization. The ensembles are created by perturbing the surface forcing fields
(wind-stress, fresh-water flux and heat flux) and the observations (temperature and salinity profiles) used in the assimilation process.
Cycled 3D-Var experiments over the period 1993-2000 are presented to test the sensitivity of the analyses to two flow-dependent formulations of the background-error standard deviations (sb) for temperature and salinity. The first formulation is based on an empirical parameterization of sb in terms of the vertical gradients of the background temperature and salinity fields, while the second formulation involves a more sophisticated approach that derives sb from the spread of an ensemble of analyses. In both experiments, the observation-error standard deviations (so) are geographically dependent and estimated from a model-data comparison prior to assimilation. An additional 3D-Var experiment that employs the parameterized sb but a simpler so formulation, and a control experiment involving no data assimilation were also conducted and used for comparison. All 3D-Var experiments produce a significant reduction in the mean and standard deviation of the temperature and salinity innovations compared to those of the control experiment. Comparing innovation statistics from the two sb formulations shows that both formulations produce similar results below approximately 150 m but the parameterized sb produce slightly better results above this depth where statistical consistency checks indicate that the ensemble sb are underestimated. The rate at which observational information is lost between cycles, however, is shown to be much reduced with the ensemble sb, suggesting that the analyses produced with the ensemble sb are in better balance than those produced with the parameterized sb. Sea surface height (SSH) anomalies in the northwest Atlantic and zonal velocities in the equatorial Pacific, which are fields not directly constrained by the observations, are clearly better with the ensemble sb than with the parameterized sb when compared to independent data. Results show that while some aspects of those variables are improved with data assimilation (SSH anomalies and currents in the central and eastern Pacific), other aspects are degraded (SSH anomalies in the northwest Atlantic, currents in the western Pacific). Areas for improving the ensemble method and for making better use of the ensemble information are discussed.
(Last Updated: 30-10-2008)