B.3 - Progress in Oceanographic Applications
(Moderator: Don Chambers)

   Don Chambers (12)
   Maik Thomas (12)
   R.M. Ponte (12)
   Andreas Macrander (12)
   Ming Fang (12)
   Discussion (10)
   Roelof Rietbroek (12)
   James Morison (12)
   Yuichi Aoyama (10)
   Discussion (10)
   Richard Ray (12)
   Ki-Weon Seo (12)
   Discussion (10)
   Nikolaos Pavlis (12)
   Nikolai Maximenko (5)
   Huei-Ping Huang (8)
   Y. Tony Song (12)
   Discussion (5)

Don Chambers (12)
Presenter: Don Chambers
Co-Authors:
Title: How to Properly use GRACE Time-Variable Data for Ocean Applications

Abstract: Because ocean mass variations are much smaller than hydrology variations, one has to be particularly careful in how the GRACE gravity coefficients are used. One cannot simply take the time-variable coefficients as released by the SDS centers and obtain the ocean bottom pressure or ocean mass variations. In this talk, I will summarize the important issues for using Release-02 data and higher. These include adding back the reference ocean model (i.e., what is the difference between GAB, GAC, GAD and which should you use), use of a geocenter model, and glacial isostatic adjustment (GIA) corrections. Just as for applications over Greenland and Antarctica, the correction for GIA over the ocean is vital if one is interested in observing trends from the redistribution of water mass between the land and ocean.

Maik Thomas (12)
Presenter: Maik Thomas
Co-Authors: H. Dobslaw
Title: Simulation and observation of regional and total ocean mass signals

Abstract: Recently reprocessed GRACE gravity fields in connection with newly developed filter algorithms have greatly improved the ability to directly map ocean bottom pressure anomalies from space on a monthly basis. Comparisons with sterically corrected Jason-1 sea level anomalies indicate that reliable ocean mass anomalies from GRACE can be now achieved on spatial averages down to 500km away from the tropics. Robust signals of up to 10hPa are identified in the North Pacific and in several regions of the Southern Ocean, what is generally confirmed by Jason-1 with correlations approaching 0.8 in these areas.

By means of numerical simulations with the global ocean model OMCT, GRACE observed mass anomalies are identified to be primarily connected to changes in barotropic transports caused by, e.g., a seasonal shift of the mean flow path of the Kuroshio or by local vorticity variations west of the Drake passage. Obviously, this leads to the opportunity to link GRACE observations directly to climate relevant circulation processes in the global ocean.

Moreover, simulations of the ocean's general circulation have been performed to estimate the impact of atmospheric and continental freshwater fluxes on global ocean mass redistributions. Applying consistent ECMWF atmospheric forcing fields, OMCT and the Hydrological Discharge Model (HDM) have been coupled via freshwater fluxes to nearly close the hydrological cycle and to assess the impact of continental discharge on near-shore mass anomalies. Comparisons of simulated time-series of total ocean mass with corresponding variations calculated from monthly mean GRACE gravity fields turn out that seasonal variations of total ocean mass are reproduced quite reasonably by the model combination ECMWF-OMCT-HDM. In addition to total oceanic mass variations, time-varying regional oceanic mass anomalies mainly caused by river runoff have been found to be mainly confined to local effects in coastal areas, except for the Arctic. High amounts of meltwater routed into the Arctic basin predominantly during July have have been found to cause a monthly mean mass anomaly of up to 1hPa regionally averaged over the Arctic ocean, large enough to be potentially detectable by GRACE.

R.M. Ponte (12)
Presenter: R.M. Ponte
Co-Authors: K.J. Quinn; P. Heimbach; C. Wunsch
Title: Seasonal cycle in ocean bottom pressure from model and GRACE estimates

Abstract: Seasonal variability in ocean bottom pressure is analyzed using the filtered GRACE fields provided by Chambers (2006, Geophys. Res. Lett., 33, L17603) and model fields from the ECCO-GODAE ocean state estimates, obtained by fitting a state-of-the-art general circulation model to most available ocean data in a least-squares sense, using best known statistics of model and data errors. Results indicate that a substantial part of the observed seasonal variability in ocean bottom pressure is associated with the annual cycle in net freshwater flux and net atmospheric pressure over the oceans. To a first approximation, these effects result in spatially constant bottom pressure signals, with little dynamical relevance. Comparison of spatially varying seasonal signals reveals moderate agreement between GRACE and model results, particularly over the Southern Ocean where strongest variability at both annual and semiannual periods is present. Phase patterns tend to agree well, although model amplitudes are generally weaker. Apart from spatial mean fluctuations, considerable uncertainty remains in both data and model estimates of bottom pressure, judging from the relative size of expected variability (order 1 cm water equivalent) and the spread among different GRACE and model estimates. Prospects are promising for the use of monthly GRACE data as constraints on the ECCO-GODAE estimates, but much more complete understanding of the data errors is required before a useful misfit function can be constructed.

Andreas Macrander (12)
Presenter: Frank Flechtner
Co-Authors: A. Macrander; C. Bšning
Title: Comparison of Ocean Bottom Pressure estimates from different GRACE products

Abstract: Based on gravity field measurements, GRACE provides estimates of Ocean Bottom Pressure (OBP) variability. In the Southern Ocean, a comparison with in-situ time series of moored bottom pressure recorders (PIES) indicates that GRACE is able to realistically observe some aspects of oceanic variability. Nevertheless, GRACE products provided by CSR, GFZ, JPL, GRGS and ITG feature significantly different patterns of spatially coherent OBP variability, which implies major differences in raw data processing, de-aliasing and other applied correction models.

Here, spatial and temporal variability of different GRACE products are analyzed and compared with in-situ measurements in the Southern Ocean and simulated OBP derived from the 3D global Finite Element Sea-Ice Ocean Model (FESOM) which was recently developed at the Alfred Wegener Institute for Polar and Marine Research (AWI).

First results indicate that major anomalies on the sub-basin scale are well captured in GRACE retrievals. However, in some regions in the equatorial Atlantic the agreement between GRACE and FESOM is rather poor.

Ming Fang (12)
Presenter: Ming Fang
Co-Authors: R. Ponte; B. H. Hager; and C. Wunsch
Title: Long Wavelength Ocean Bottom Pressure: a Bridge between Hydrology and Oceanography

Abstract: Annual variability in surface mass redistribution, according to GRACE inversions, is in the range of 30 cm equivalent water column in terms of basin wide averages. This basin-wide load results in elastic variations in the elevation of the surface of the solid earth by about 3 cm in its near field. Sea floors in the vicinity of major water storages are deformed by roughly the same amount. In addition, changes in the gravitational potential over the oceans caused by hydrologic mass variations on land exceed 1 cm over large areas. These important sources of variability in ocean thickness and hence in the bottom pressure (BP) field, have not been accounted for up to now in ocean general circulation models (OGCM). Technically, the BP perturbation relative to an OGCM can be rephrased as a problem of perturbation in sea level measured against the deformed seafloor under the constraint of total mass conservation between the land and sea. In this framework, the deformed sea floor, the deformed equipotential at the sea surface, and the surface mass exchange between the land and sea are all taken into account. The BP field is solved from a full self-gravitating sea level equation with the variation in mass over land mass (inverted from GRACE with our wavelet filter) and the "imperfect" BP field (created by the MIT OGCM) as the input signals. The theory is similar to what we have employed recently in recovering the long wavelength GRACE geoid over the ocean, except that the effect of deformed sea floor is treated as second order term in perturbing the long wavelength GRACE geoid, while in perturbing the BP field, the deformed sea floor comes out as a first order term. Based on availability of the required data, we consider in this report a two-year period from Jan. 2003 to Dec. 2004. The sea level equation is solved by a semi-analytical method developed earlier in our group based on truncated spherical harmonics. The accuracy of the results depends upon the accuracy of the land mass variations inverted from GRACE. Tests of our inversion using the wavelet filter are done based on inversions of the synthetic geoid created by the GLDAS global hydrology model. The resulting BP variations will be compared to the "imperfect" BP field produced by the OGCM.

Discussion (10)

Roelof Rietbroek (12)
Presenter: Ejo Schrama
Co-Authors: B. Wouters; P. LeGrand; E.J.O. Schrama
Title: GRACE validation with bottom pressure data in the Crozet-Kerguelen region

Abstract: Two time series of deep ocean bottom pressure records (BPRs) in between the Crozet Islands and Kerguelen are compared with GRACE (Gravity Recovery And Climate Experiment) equivalent water heights. An analysis of the correlation is performed for four time series: 1) monthly averages of the equivalent water height at the Crozet Islands, 2) the same near the Kerguelen Islands, 3) the mean of the two preceding series and 4) the difference between the two locations expressed in terms of geostrophic transport. We find that smoothed GRACE solutions are strongly correlated with the BPR data of the first three series. Correlation coefficients in the order of 0.6-0.8 are found for spatial smoothing radii of around 800 km. Consequently GRACE measures real oceanic mass variations in this region. The good agreement appears to be due to dominant large scale signals, such as seasonal signals and variations in the Antarctic Circumpolar Current. The fourth series representing geostrophic transports shows a weaker resemblance due to the spatial correlation between the BPR stations, associated with the smoothing process. Future improvements in increasing spatial resolution of GRACE data promise the retrieval of deep ocean currents.

James Morison (12)
Presenter: James Morison
Co-Authors: John Wahr, Ron Kwok, Cecilia Peralta-Ferriz
Title: GRACE and In Situ Bottom Pressure Measurements Reveal Change in the Arctic Ocean

Abstract: In the late 1980s and through the 1990s we saw major shifts in the Arctic Ocean hydrography. The influence of Atlantic Water in the Arctic Ocean became more widespread and intense and the pattern of water circulation and ice drift shifted, resulting in a more cyclonic circulation. These changes became manifest in the central Arctic near the North Pole as increases in upper ocean salinity and Atlantic Water temperature. They occurred in concert with a decrease in surface atmospheric pressure. With the aim of helping to track such changes, we have undertaken in situ ocean bottom pressure measurements and the analysis of Gravity Recovery and Climate Experiment (GRACE) data. For the in situ measurements, with the help of engineers at the NOAA Pacific Marine Environmental Lab, we have developed an Arctic Bottom Pressure Recorder (ABPR), which is suitable for deployment through pack ice. The ABPRs are equipped with acoustic modems to allow annual data recovery while leaving the instruments undisturbed on the bottom for up to 3 years. The first year of data from gauges installed near the North Pole was recovered in April 2006. The comparison between GRACE-derived bottom pressure at the North Pole and the ABPR data is quite good. The GRACE data are filtered with a 400 km radius Gaussian filter, so their footprint easily covers both ABPR locations. At this scale, the two ABPR records are highly correlated. Both GRACE and the ABPRs show a declining bottom pressure trend in 2005-2006. The complete GRACE record indicates a -2.43 cm/yr trend from 2002 to 2006. We believe this trend is associated with a steric change due to a drop in upper ocean salinity near in the central Arctic Ocean as tracked for the last 6 years by the North Pole Environmental Observatory (NPEO). We have also investigated the effect on bottom pressure of a hypothetical return to pre-1990s hydrography over a larger area of the Arctic Ocean. We find correspondence with the spatial distribution of bottom pressure trends from GRACE, especially a decrease in bottom pressure in the Makarov Basin associated presumably with the return of less saline, Pacific-derived upper ocean water to that region. The differences between the GRACE bottom pressure trends and the pressure trends due to the hypothetical change in density yields a pattern of sea surface height trends consistent with observed trends in ice drift towards a anticyclonic pattern similar to the pre-1990s state.

Yuichi Aoyama (10)
Presenter: Yuichi Aoyama
Co-Authors: K. Doi; K. Shibuya; Y. Nogi
Title: Comparison of non-tidal ocean bottom pressure data with GRACE-derived equivalent water thickness off Lutzow-Holm Bay, Antarctic Ocean

Abstract: Several precise geodetic observations with VLBI, GPS, absolute gravimeter (AG), superconducting gravimeter (SG), etc. are continued around the Syowa Station area, Antarctica, for more than seven years. In addition, we have also conducted GPS campaign measurements in the bedrock areas around Syowa Stations and ocean bottom pressure (OBP) measurements off Lutzow-Holm Bay, Antarctic Ocean with objective of studying Earth's responses (e.g., crustal uplift, etc.) to geodynamics (e.g., mass redistributions) in this area. These observed data are useful to calibrate GRACE data. Practically, we have started to make comparisons between SG and GRACE-derived gravity fields (Doi et al., AGU Fall Meeting) and between OBP and a GRACE-derived equivalent water thickness.

The OBP gauges were installed to the north of Syowa Station at (37.8E, 66.9S). We have already obtained about 1.2 years OBP data from Dec. 2004 to Feb. 2006. The observed OBP data at 1 minute sampling intervals is decomposed by tidal analysis software BAYTAP into tidal (M3 to SSA), trend and short-term irregular components. After removal of an instrumental initial drift by fitting exponential function, the trend component is adapted by low-pass filter. This smoothed non-tidal residual is compared with the equivalent water thickness that is available from Univ. of Colorado Real-Time GRACE Data Analysis Site. The result shows their rough agreement. We plan to continue OBP measurements for several years and reveal the changes in the water mass distribution around this area by applying geodetic observations and GRACE data.

Discussion (10)

Richard Ray (12)
Presenter: Richard Ray
Co-Authors: G Egbert, S Erofeeva, S Luthcke, D Rowlands
Title: Tides and GRACE

Ki-Weon Seo (12) Presenter: Ki-Weon Seo
Co-Authors: C. R. Wilson; S. Han; D. E. Waliser;
Title: GRACE aliasing error from ocean tides

Abstract: GRACE aliasing errors associated with ocean tides are simulated from Aug. 2002 to Feb. 2006. Differences between GOT00.2 and TPXO6.2 tide models are used to estimate errors of 8 tidal constituents (Q1, O1, P1, K1, N2, M2, S2 and K2). P1, K1, S2 and K2 tides are found to pose the greatest problems in GRACE monthly solutions because their aliased periods are longer than a month. We show how aliasing errors from ocean tides corrupt mass change estimates over land even though the errors are over the oceans. These errors may affect geophysical interpretation of GRACE estimates at high latitudes, for example, ice mass budgets over Antarctica. Furthermore, we show that the slow decay in the GRACE orbit causes tidal component aliases to differ from pure frequencies as the orbit evolves.

Nikolaos Pavlis (12)
Presenter: Nikolaos Pavlis
Co-Authors: S.A. Holmes
Title: Dynamic Ocean Topography Estimates Using GRACE-Based Gravitational Models

Abstract: A new Mean Sea Surface (MSS) model is under development at the Danish National Space Center (DNSC). This MSS model incorporates data from several altimeter missions (Geosat GM, ERS-1 and ERS-2, TOPEX/Poseidon and Jason-1, GFO, Envisat, and ICESat). In parallel, a new Preliminary Gravitational Model (PGM) has been developed recently, complete to spherical harmonic degree and order 2190. This model is an intermediate solution leading to a new Earth Gravitational Model complete to degree and order 2190. We have used a preliminary version of the new MSS from DNSC and the latest PGM, to estimate the mean Dynamic Ocean Topography (DOT). Using the same MSS, we have also computed the DOT implied by the GGM02C and the EIGEN-GL04C models. We will present and compare the various DOT estimates, both against each other and against the ECCO estimate of the DOT. Of particular interest in these comparisons is the treatment of systematic errors in GRACE-based gravitational solutions that manifest themselves as ÔstripesÕ, predominantly in the North-South direction. We will discuss our viewpoint regarding the advantages and disadvantages of various Ôde-stripingÕ approaches.

Nikolai Maximenko (5)
Presenter: Don Chambers (for Nikolai)
Co-Authors:
Title: Adding in situ measurements to altimeter/GRACE dynamic topography

Abstract:

Huei-Ping Huang (8)
Presenter: Alexey Kaplan
Co-Authors: A. Kaplan, E.N. Curchitser
Title: Application of GRACE data to investigations of new aspects of mid-ocean currents

Abstract: Satellite surveys of global oceans have greatly enhanced our knowledge of the structures of ocean circulations. Among the notable recent findings are the alternating mid-ocean zonal currents derived from satellite altimetry. While the altimetry data provides the detailed structures of the surface zonal currents for the anomalous velocity field, an accurate determination of the time-mean absolute velocity field is more challenging. Recent attempts of combining GRACE-based geoid with altimetry data point to a possible new direction for constructing the absolute currents. To assess the usefulness and limitations of this approach, a preliminary comparison is made between the GRACE & altimetry based absolute velocity field and that simulated with an eddy-resolving model for the Pacific Ocean. The model is shown to simulate multiple zonal bands in the zonal velocity fields at both surface and 1000 m depth. The resolution of the GRACE & altimetry based absolute velocity field is limited by the effective resolution of the GRACE-based geoid, which is on the order of 100 km compared to 10 km for the ocean model. As such, fine structures of multiple zonal bands that are evident in the latter are not well-resolved in the former. The large-scale structures of the absolute velocity field are otherwise similar in both satellite observation and model simulation. The implications of these results and future directions of maximizing the uses of GRACE data for the investigations of mid-ocean currents will be discussed.

Y. Tony Song (12)
Presenter: Y. Tony Song
Co-Authors: Ki-Weon Seo, Victor Zlotnicki
Title: Estimation of interbasin transport using GRACE-observed OBP

Abstract: Here we propose a method, the combination of the "geostrophic control" formula of Garrett and Toulany [1982] and the "hydraulic control" theory of Whitehead et al. [1974], allowing the use of satellite-observed sea-surface-height (SSH) and ocean-bottom-pressure (OBP) data for estimating interbasin transport. The new method also allows separating the interbasin transport into surface and bottom fluxes that play an important role in maintaining the mass balance of the regional oceans. Comparison with model results demonstrates that the combined method can estimate the seasonal variability of the strait transports and is significantly better than the method of using SSH or OBP alone.

Discussion (5)