A. Process Study Title
Southern Ocean Fluxes
B. Scientific Leader
Prepared by K. Speer for CLIVAR Southern Ocean Working Group
C. Beginning/End Date
2004/2010
D. Panel/Working Group
Southern Ocean Working Group
1. Objectives and Relevance
1.1 The general strategy to enable CLIVAR to obtain good surface flux fields has been to rely on: 1) in-situ measurements directed at improving flux parameterizations, calibration, and validation, and at anchoring flux fields, 2) sensor development and calibration, 3) development of remote sensing-based air-sea flux fields, and 4) development of air-sea flux fields from mixed data sets (in-situ, numerical weather prediction (NWP) model fields (Fig 1.), and remote sensing data.
The objectives are to provide measurements and model output to characterize heat, precipitation, evaporation, winds, and various gas exchanges over open-ocean and sea-ice regions in the Southern Ocean.
Figure 1 illustrates some of the discrepancies among existing flux products. Many problems with both ship-based observations and numerical weather center products are well known and documented for northern hemisphere and equatorial sites, but the density of observations in the Southern Ocean is low, and corresponding parameter ranges have been less explored. It is not clear that corrections from other sites can be carried over without further study. An outstanding question and key goal is to determine if the Southern Ocean is a region of net heat gain or loss.
Figure 1: Comparison of heat flux components between re-analyses and ship-based climatologies, zonally averaged over the southern hemisphere. Major biases result from different radiative components and bulk formula differences (From: Grist, J. P. and Josey, S. A., 2002: Inverse analysis of the SOC air-sea flux climatology using ocean heat transport constraints, Journal of Climate, in prep.)
1.2 Air-sea fluxes are of fundamental importance to climate studies because they provide the basis for understanding the coupling between the ocean and atmosphere systems. Moreover, the air-sea flux distribution is at least diagnostically related to the accumulation of anthropogenic CO2 in the ocean (Fig.1). The accumulation over the 40-50S band is larger than in any other latitude band, including the North Atlantic. Climate models suggest that the uptake and accumulation of CO2 in the Southern Ocean will decrease as a result of circulation changes driven by enhanced greenhouse warming.
Figure 2: Annual mean air-sea flux of CO2 (Takahashi et al. 2002). Strong sinks appear in the Southern Ocean, especially over mode waters.
2. Process Study Plans
2.1 Overall Approach
The approach is to encourage a number of projects that validate existing products and improve the observational base and the incorporation of observations into the global operational weather forecasting network. The ignorance of the operational centers to existing southern hemisphere sea level pressure data has been documented. Remote sensing will play an important role.
Constraints on air-sea flux fields from estimates of heat divergence (between two ocean hydrographic sections, for example), by ocean data assimilation (the best "fit" of an ocean model and observations), and by atmospheric model heat divergences are also important approaches since they do offer some independent information. Careful attention to circulation and to internal ocean mixing uncertainties is needed to exploit these methods for air-sea fluxes.
Validation. Air-sea fluxes from various sources have been compared, primarily based on shipboard measurements of the raw meteorological variables and subsequent modeling for bulk fluxes. Observed biases between NCEP latent heat flux and observations, for example, over the global ocean are linked to choices of bulk model parameters. Special problems exist when ice is present, and this is a crucial point for Southern Ocean fluxes and deep water formation since such a large area is covered by sea-ice during the winter months.
Flux validation data exist due to the efforts made during WOCE and other experiments (Fig. 3; note: the SEAFLUX website has records of WOCE data only in the Southern Ocean). These data require further exploitation to resolve errors and biases on a regional basis.
WOCE Surface Meteorology Data Coverage
Figure 3: WOCE ship tracks and frequency of meteorological measurements (from the WOCE surface flux SAC at COAPS).
Fluxes of CO2 (Fig. 2), oxygen, and other gases merit inclusion in air-sea flux studies since they are important for climate change, either directly or via biological processes.
Winds are a key variable for almost all air-sea fluxes. Validation studies have demonstrated significant differences in average quantities, but remote sensing of wind fields is now allowing for corrections to be made to the higher frequency and smaller spatial scale spectrum of variability (Fig. 4).
Figure 4: Scatterometer corrected surface pressure compared to the NCEP reanalysis. The low-pressure system south of Tasmania is absent in NCEP (Bourassa, FSU).
To organize the approach the project is broken into 3 phases:
Phase 1) Diagnostic and validation components
Phase 2) Observational components
Phase 3) Synthesis and analysis
These phases are not necessarily sequential; validation will be important throughout the project. But they reflect an emphasis to guide budgeting priorities as the project proceeds. An initial goal will be to assemble all the relevant Southern Ocean air-sea flux validation data (from cruises and past process studies) as a community resource.
The observational plan is to be developed in concert with other programs and ongoing work (e.g. RIME). One observational component (profiling floats in Antarctic sea-ice regions) is presented separately. Another connected to a project to study water mass formation in the Southeast Pacific is also submitted separately. Further components require coordination with other working groups.
2.2: Time Line.
2004/2006 Phase I Diagnostic components
2006/2008 Phase II Observational components
2008/2010 Phase III Synthesis and analysis
Note: Model activities parallel the above
2.3 Agency programs likely to provide support:
NOAA, NSF, NASA, DOE, …
3. Observational Elements
1) Large-scale flux divergences
2) Expanded buoy observations of basic met variables
3) IMET buoy times series for 1-2 years
4) Remote sensing for air-sea flux variables and fluxes
5) Surface boundary layer budgets (ocean and atmosphere)
3.1 Linkages to broad-scale observing efforts for the ocean, atmosphere, land and cryosphere. This is a broad-scale effort.
3.2 Regional enhancements/enhanced monitoring. More realistic wind stress, heat and fresh-water budgets are expected to contribute to a better understanding of the physics of ocean, through improved modeling results and diagnostic studies of flow and stratification. Fresh water and ice budgets in regions of sea-ice are severely compromised by the lack of validation and error estimates.
Several locations have been proposed for the IMET buoy, including one south of Tasmania on a regularly occupied line, and another in the Bellingshausen Basin. Discussions with the CLIVAR time-series group are ongoing.
3.3 Estimated cost per year for each key observational element (do not include costs for observational systems already in place).
N.B. Only Phase I Diagnostic component is budgeted (based on support staff requirements, not PI salaries). Observational components will be budgeted once priorities are set by US CLIVAR and the Southern Ocean Working Group.
Year 1 2 3 4 5 Total
04-05 05-06 06-07 07-08 08-09
Personnel 350 350 250 250 250 1450
Equipment 50
Travel/Other 20 10 10 20 10 70
Total 420 360 260 270 260 1570
4. Modeling Elements
Essentially operational numerical weather prediction products.
4.2 Required resources. Access and data storage capacity.
5. Feasibility and Readiness
Validation studies are ongoing; observational plans are at various stages of readiness.
6. National and international links and partners
To be determined.