Back to Peter Oke’s HOMEPAGE

Bluelink ReANalysis - BRAN

Publications

link to a complete list of publications

Schiller, A., K. R. Ridgway, C. R. Steinberg, and P. R. Oke, 2009: Dynamics of three anomalous SST events in the Coral Sea. Geophysical Research Letters, doi:10.1029/2008GL036997, in press.

Oke, P. R., G. B. Brassington, D. A. Griffin and A. Schiller, 2008: The Bluelink Ocean Data Assimilation System (BODAS). Ocean Modelling, 20, 46-70, doi:10.1016/j.ocemod.2007.11.002.

Schiller, A., P. R. Oke, G. B. Brassington, M. Entel, R. Fiedler, D. A. Griffin, and J. Mansbridge, 2008: Eddy-resolving ocean circulation in the Asian-Australian region inferred from an ocean reanalysis effort. Progress in Oceanography, doi:10.1016/j.pocean.2008.01.003.

Oke, P. R., and A. Schiller, 2007: Impact of Argo, SST and altimeter data on an eddy-resolving ocean reanalysis. Geophysical Research Letters, 34, L19601, doi:10.1029/2007GL031549.

Oke, P. R., A. Schiller, G. A. Griffin, G. B. Brassington 2005: Ensemble data assimilation for an eddy-resolving ocean model. Quarterly Journal of the Royal Meteorological Society, 131, 3301-3311.

GODAE 2008 Poster: The Bluelink analysis and re-analysis systems at work in Australia, Poster (link [pdf, 4MB])

GODAE 2008 Poster: Dynamics of a monster eddy (link [pdf, 5MB])

Ocean Forecasting Australia Model – OFAM

OFAM1.0

OFAM is based on version 4.0d of the Modular Ocean Model (Griffies et al. 2004), using the hybrid mixed layer model described by Chen et al. (1994). OFAM is intended to be used for reanalyses and short-range prediction. The horizontal grid has 1191 and 968 points in the zonal and meridional directions respectively; with 1/10 degree horizontal resolution around Australia (90-180E, south of 17). Outside of this domain, the horizontal resolution decreases to 0.9 degrees across the Pacific and Indian basins (to 10E, 60W and 40N) and to 2 degrees in the Atlantic Ocean. OFAM has 47 vertical levels, with 10 m resolution down to 200 m depth. The topography for OFAM is a composite of topography sources including dbdb2 and GEBCO. Horizontal diffusion is zero. Horizontal viscosity is resolution and state-dependent according to the Smagorinsky viscosity scheme (Griffies and Hallberg 2000).

OFAM is described in more detail by Oke et al. (2008) and Schiller et al. (2008).

OFAM2.0

OFAM is based on version 4.1d of the Modular Ocean Model (Griffies et al. 2004), using the hybrid mixed layer model described by Chen et al. (1994). OFAM is intended to be used for reanalyses and short-range prediction. The horizontal grid has 1191 and 968 points in the zonal and meridional directions respectively; with 1/10 degree horizontal resolution around Australia (90-180E, south of 17). Outside of this domain, the horizontal resolution decreases to 0.9 degrees across the Pacific and Indian basins (to 10E, 60W and 40N) and to 2 degrees in the Atlantic Ocean. OFAM has 47 vertical levels, with 10 m resolution down to 200 m depth. The topography for OFAM is a composite of topography sources including XXX XXX Horizontal diffusion is zero. Horizontal viscosity is resolution and state-dependent according to the Smagorinsky viscosity scheme (Griffies and Hallberg 2000).

The topography in OFAM2.0 is different to that of OFAM1.0. The differences are demonstrated in the Figures below.

Bluelink Ocean Data Assimilation System (BODAS)

BODAS used an ensemble optimal interpolation (EnOI) scheme that uses a stationary ensemble of intraseasonal model anomalies, obtained from a non-assimilating model run. Observations that can be assimilated by BODAS include along-track SLA (atSLA) from altimeters and tide gauges, in situ T and S observations and satellite SST. In principal, BODAS can assimilate any type of observation of temperature, salinity, sea-level or velocity.

BODAS is described in detail by Oke et al. (2008); see also BODAS link.

BRAN

BRAN is a multi-year integration of OFAM that assimilates observations using BODAS. The goal of BRAN is to provide a realistic quantitative description of the three-dimensional time-varying ocean circulation of all physical variables (temperature, salinity, sea-level and three components of velocity) for the last few decades. The following BRAN experiments have been performed:

SPINUP1: suffered from a warm bias and a trend that has been traced to an error in the surface fluxes.
BRAN1.0: the first version of BRAN assimilated along-track sea-level anomalies (SLA) from altimeters, in situ temperature and salinity (from Argo, XBT, TAO etc). BRAN1.0 spanned the period October 1992 to December 2004. Results from BRAN1.0 are described by Oke et al. (2005), demonstrated the potential of the Bluelink system to realistically represent the mesoscale ocean circulation. BRAN1.0 suffered from a warm bias, as in SPINUP1.
SPINUP4/5: tuning of the mixed-layer scheme and corrections to the surface fluxes significantly improved the simulation; the warm bias and trend are not present.
BRAN1.5: the second version of BRAN assimilated along-track SLA, SST from AMSR-E and in situ temperature and salinity. BRAN1.5 spanned the period January 2003 to June 2006. BRAN1.5 realistically reproduces the mesoscale circulation around Australia with errors of ~5-10 cm for sea-level, ~0.6^o for SST; <1^o for sub-surface T and <0.15psu for salinity. Results are described by Oke et al. (2008).
BRAN2.1: the third version of BRAN uses a very similar configuration as BRAN1.5 that is described by Oke et al. (2008). BRAN2.1 also assimilated Pathfinder SST before 2002, when AMSR-E observations became available. Results from BRAN2.1 are described by Schiller et al. 2008).
BRAN2.2: like BRAN2.1 except in the way the model is initialised. The latest series of test runs use incremental analysis updating, where U, V, T, S and eta are explicitly updated over a number of time-steps. These experiments provide smooth, realistic fields that fit the observations better than all previous experiments.

Table 1: Completed experiments using OFAM1.0.

*BRAN1.0*	*SPINUP4-5*	*BRAN1.5*	*BRAN2.1*	*BRAN2.2*
10/1992-12/2004	SP4: 1/1997-12/2005 SP5: 1/1992-12/1996	1/2003-6/2006	10/1992-12/2006	12/2006-4/2008
Assimilates along-track SLA, in situ T & S	no assimilation	Assimilates along-track SLA, in situ T & S, AMSRE-SST	Assimilates along-track SLA, tide-gauge SLA, in situ T & S, AMSRE or Pathfinder SST	Assimilates along-track SLA, tide-gauge SLA, in situ T & S, AMSRE
no rivers	no rivers	no rivers	Seasonal climatological river fluxes	Seasonal climatological river fluxes
SSS restoring (30 days); SST restoring (30 days)	SSS restoring (30 days); SST restoring (30 days)	no SSS or SST restoring	SSS restoring (30 days in deep water only); no SST restoring	SSS restoring (30 days in deep water only); no SST restoring
ECMWF surface heat, freshwater and momentum fluxes	ECMWF surface heat, freshwater and momentum fluxes	ECMWF surface heat, freshwater and momentum fluxes	ECMWF surface heat and freshwater fluxes; and momentum fluxes from 10 m winds	ECMWF surface heat and freshwater fluxes; and momentum fluxes from 10 m winds
3 day assimilation cycle	no assimilation	7 day assimilation cycle	7 day assimilation cycle	7 day assimilation cycle
Error in surface fluxes; a few bugs in BODAS	n/a	Some in situ temperature observations processed incorrectly	Error in topography in Torres Strait and Cook Strait	Error in topography in Torres Strait and Cook Strait
72-member ensemble based on SPINUP1	n/a	72-member ensemble based on SPINUP4/5	120-member ensemble based on SPINUP4/5	120-member ensemble based on SPINUP4/5
Updates U, V, T, S and eta in single step	n/a	Updates T, S and eta via nudging over 1-day.	Updates T, S and eta via nudging over min(1-day, local inertial period).	Updates T, S, U , V, and eta via incremental analysis updating applied over 12 hours.
Fields are noisy, often realistic, but under-fit the observations.	Sea-level falls because of the neglect of rivers	Fields are smooth and realistic, but under-fit observations.	Fields are smooth and realistic, but under-fit observations.	Fields are smooth and realistic, with better fit to observations than BRAN2.1 (still under-fitting Argo T/S)

Table 2: Completed experiments using OFAM2.0.

*SPINUP6p8*	*BRAN3p0*	*BRAN3p1*
1/1992-12/2008	1/2003-12/2009	1/2003-12/2009
no assimilation	Assimilates along-track SLA, tide-gauge SLA, in situ T & S, AMSRE and 4 km Pathfinder SST and 4 km NAVO SST	Assimilates along-track SLA, tide-gauge SLA, in situ T & S, AMSRE and 4 km Pathfinder SST and 4 km NAVO SST
Seasonal climatological river fluxes	Seasonal climatological river fluxes	Seasonal climatological river fluxes
SSS restoring (30 days in deep water only); no SST restoring	SSS restoring (30 days in deep water only); no SST restoring	SSS restoring (30 days in deep water only); no SST restoring
ERA-interim surface heat, freshwater and momentum fluxes	ERA-interim surface heat, freshwater and momentum fluxes	ERA-interim surface heat, freshwater and momentum fluxes
n/a	4 day assimilation cycle	4 day assimilation cycle
n/a	Error in MSL – impacts atSLA assimilation Error in freshwater fluxes – not globally conservative Poor ensemble – daily mean minus month mean (formally used 3-day mean minus 3-month mean) – makes length-scales shorter – and increments more noisy Possibly bug in BODAS (?)	No known errors … yet
n/a	144-member ensemble based on SPINUP6p8	144-member ensemble based on SPINUP6p8 (3-day mean minus 3-month mean)
n/a	Updates T, S, U , V, and eta via adaptive nudging (Sandery et al. 2011) applied over 24 hours.	Updates T, S, U , V, and eta via adaptive nudging (Sandery et al. 2011) applied over 24 hours.
	Some pulsing in sea-level due to error in MSL Some mean in increments due to error in surface fluxes.	Almost perfect J

Results

Graphics archive of BRAN experiments