Newsletter : December 2001

Sunday, September 12 2004 @ 08:42 PM EDT

A lot of news to report; thanks to all of you who have provided input. Please keep sending your input.

Re: the SEAFLUX web page. I am up to date on posting data. However, I am somewhat behind in organizing some parts of the web page and getting things posted, hope to rectify this shortly. When somebody sends me a copy of a submitted paper of figures, I am sometimes unsure as to whether it is OK to post these on the web page. If you see something posted on the web page that you don't want there (or vice versa), please let me know.

Thanks to all who provided input for the SEAFLUX talk at the GEWEX Radiation Panel meeting last month. The Panel was very pleased with the progress on SEAFLUX. Arnie Gruber of the GPCP group (precip) said that he would produce a special high-resolution rainfall data set for the 1999 year (which is the designated year for our global intercomparison). There was considerable discussion on the interactions with the WCRP/SCOR Working Group on Radiative Fluxes. The next SEAFLUX presentation to the WCRP will be at the JSC meeting in March (Hobart, AU).

From Sue Chou: Version 2 of Goddard Satellite-Based Surface Turbulent Fluxes (GSSTF) dataset, derived from SSM/I data (July 1987 to December 2000; 1-degree lat by 1-degree long) is archived and distributed by the Goddard DAAC. (http://daac.gsfc.nasa.gov:82/CAMPAIGN_DOCS/hydrology/hd_main.html). This is linked to the SEAFLUX web site from the links page.

From Bin Lin: New TMI-based flux data set for the tropics is available. The data is estimated from TMI measurements during 01/98 ~ 08/98 (the CERES TRMM time period), covering -180deg ~ +180deg lon & 30S ~ 30N lat. The temporal and spatial resolutions are 1 h and 1 deg grid, respectively. We have made an initial comparison for zonal mean fluxes with GSFC's SSM/I retrievals. The two results are very close. This data set is available from the SEAFLUX page, Intercomparison, Satellite and NWP Products, Lin et al. satellite fluxes, SEAFLUX/SATELLITE/TMI/Lin

From Abderrahim Bentamy: A new flux product is available based on the ERS-2, NSCAT, SSMI, and AVHRR. The product is global, weekly, 1o, for the period 30 September 1996 through 29 June 1997. The product and documentation can be obtained the SEAFLUX page, Intercomparison, Satellite and NWP Products, Bentamy et al. satellite fluxes.

From Joerg Schulz: The HOAPS group had a small workshop at the MPI in Hamburg where they discussed the outline of HOAPS-II. The release date target is 30. June 2002. Among several adjustments to algorithms, the main new features will be: all SSM/Is are used, bulk parameterization will be COARE 2.6, additional parameters like water vapour content, liquid water content will be included, and all data will be available on 0.25x0.25 daily grids as well as orbit oriented for each platform to enable people to use this data for regional case studies.

A new data set in the Labrador Sea (Feb/Mar '97) with direct turbulence fluxes has been made available by Karl Bumke and Ian Renfrew. This data set is labeled "Labsea-Knorr" on the Intercomparison page.

Chris Fairall reports that a paper has been completed on the new version of the COARE bulk flux model, which is available at ftp://ftp.etl.noaa.gov/et7/users/cfairall/bulkalg/jclim_paper/

Michael Brunke has submitted the paper on the intercomparison of the flux models with some direct turbulence measurements from the ETL group and the French group. A draft can be viewed at: http://www.atmo.arizona.edu/personalpages/brunke/paper2/

From Will Perrie: As part of the GoMOOS (Gulf of Maine Ocean Observing System) program (October '01 - April '02), SAR winds will be collected for highest seastate conditions (~15 images) in the Gulf of Maine for comparison with buoy winds and other satellite data. Similar comparisons were conducted for Atlantic hurricanes in 2001. A high-resolution merged wind product is being developed for the North Atlantic.

From Mark Bourassa and Dudley Chelton: Differences between their scatterometer wind products have been further discussed. The problem they've been addressing is related to observational sampling, spatial and temporal variability in uncertainty, and information content (including interpretation of that information). Observational coverage from polar orbiting satellites is spatially and temporally non-uniform. This sampling results in non-uniform uncertainty in space and time, which can be reduced by averaging over larger space and time scales; however, this averaging can only be done at the expense of 'small' scale features (the scales corresponding to 'small' depend on the coverage). The inability to resolve these small-scale features uniformly in space and time leads to spatially and temporally inhomogeneous mapping errors. In extreme cases, this undersampling (combined with a gridding technique) can result in seriously erroneous results in high resolution products. For the vast majority of applications, both agree that wind (or flux) products should be smoothed sufficiently to achieve spatially and temporally homogeneous mapping errors. There will be meridional variability in uncertainty due to changes in sampling of polar orbiting satellites, but the zonal variations of the mapping errors should be made small. These considerations will be applied to our future products. Those needing higher resolution proucts need to be aware of the consequences.

Some recent papers that you might want to look at (or contact authors about submitted papers):

• Bentamy A., K. Katsaros, W. Drennan, and E. Forde 2001: Daily surface wind fields produced by merged satellite data. Gas Tranfer at Water Surfaces, (Eds.: M.A. Donelan, W.M. Drennan, E.S. Saltzman and R. Wanninkhof), Am. Geophys. Union.
• Bentamy A., N. Grima, and Y. Quilfen, 1999 : Validation of the gridded weekly and monthly wind fields calculated from ERS-1 scatterometer wind observations. Global Atmos. Ocean Sys., 6, 373-396.
• Bentamy A., P. Queffeulou, Y. Quilfen, and K. Katsaros, 1999 : Ocean surface wind fields estimated from satellite measurements. Trans. Geosc. Rem. Sens., Vol 37, No5.
• Bentamy A., K. Katsaros, A. Mestas-Nuņez, W. Drennan, E. Forde, and H. Roquet,: Remotely sensed estimates of momentum and latent heat fluxes, submitted to Journal of Climate.
• Bentamy A., P. Flament, and Y. Quilfen, 2001 : Scatterometer wind field characteristics - a new release, submitted to Canadian Journal of Remote Sensing.
• Brunke, M.A., C.W. Fairall, X. Zeng, Which Bulk Aerodynamic Algorithms Are Least Problematic in Computing Ocean Surface Turbulent Fluxes? Submitted http://www.atmo.arizona.edu/personalpages/brunke/paper2/
• Bumke, K., U. Karger and K. Uhlig: Measurements of turbulent fluxes of momentum and sensible heat over the Labrador Sea. J. Phys. Oceanography 32, 2002, accepted for publication.
• Fairall et al. (paper on new version of coare bulk flux algorithm) ftp://ftp.etl.noaa.gov/et7/users/cfairall/bulkalg/jclim_paper/
• Grimas N, A Bentamy, Y. Quilfen, 1999: Sensitivity of an Oceanic General Circulation Model Forced by Satellite Wind Stress Fields. JGR, Vol. 104, No C4, 7967-7989.
• Kubota, M. et al. 2001: Japanese Ocean Flux Data Sets with Use of Remote Sensing Observations (J-OFURO)" Journal of Oceanography, in press.
• Kubota, M. et al. 2001: Intercomparison of Various Surface Latent Heat Flux Fields, J. Climate submitted.
• Perrie, W., and P. Vachon, (high-resolution gridded vector wind fields for the Northwest Atlantic), Can. J. Remote Sensing, submitted.
• Quilfen Y., A. Bentamy, P. Delecluse, K. Katsaros, J. Gourrion, and N. Grima, 2000, Prediction of sea level anomalies measured by TOPEX/Poseidon and derived from an ocean model forced by scatterometer wind-stress fields. Int. J. Remote Sensing.
• Renfrew I.A., G.W.K. Moore, P.S. Gguest and K. Bumke: A comparison of surface-layer, surface heat flux and surface momentum flux observations, submitted.
• Renfrew, I. and GWK Moore An assessment of surface turbulent heat fluxes from the NCEP reanalysis over the western boundary currents. J. Climate, submitted.
• Weill, A., L. Eymard, et al., Determination of turbulent air-sea fluxes form various mesoscale experiments: a new perspective. J. Climate, submitted.

Here is the list of volunteers and potential coauthors to help with the BAMS article. If anyone else is interested in participating please let me know. To a certain extent I am trying to organize the figures and other contributions around what has been proposed.

• Bentamy
• Bourassa
• Brunke
• Castro
• (Chelton)
• Chou
• Clayson
• Emery
• Eymard
• Fairall
• Kubota
• Perrie
• Renfrew
• Rossow
• Schulz
• Webster
• Zeng

In the next week or so, I will circulate extended outline and some text to the participants in the BAMS article (i.e. further discussions on the paper will not be to the whole mailing list). I am thinking of figures that would show collaboration and intercomparison, rather than individual participants' work. Note I think it is wise at this point to avoid a "beauty contest" between the various satellite flux products, since we would not want any potential users to make a choice among products based on an incomplete analysis. I think it is ok to do "beauty contest" for flux algorithms, sst, winds at this point. Here are some of the intercomparison projects and figures I am thinking of, please send your comments.

1. Make the point that NWP analyses are not adequate for high resolution flux products, and that satellite can do better than NWP (in both tropics and mid/high lat):
   
   
   1. Compare Labrador Sea (KNORR) time series of latent heat fluxes with NCEP, ECMWF (note Ian Renfrew has already done this for KNORR) plus with any satellite flux products. Again need to avoid the "beauty contest" thing. Main thing is hopefully to show that satellite can do at least as well as NWP in this difficult environment. Maybe GSSFT2 is only one with high enough resolution? Note, during storms NWP fluxes get way too high, apparently because of specified surface temperature that doesn't cool in response to the storm. This case helps make point about need for high-res SST product; also, hopefully will show satellite fluxes doing better than NWP fluxes in the storms
      
      
   2. Compare instantaneous LH flux field for a period in TOGA COARE when satellite captures fluxes from a storm that was totally missed by ECMWF. I think I remember Carol Anne Clayson showing something like this at the first SEAFLUX Workshop?? This used the Curry et al. COARE product, but would be happy to use anyone else's pixel level product for the tropical pacific (need date from Carol Anne).
      
      
2. Zonal LH flux climatologies. Some version of the Kubota analysis showing annual avg zonal LH flux climatology for any global product we have (Kubota used HOAPS, GSSFT2, J-OFURO) plus perhaps any tropical only products that cover at least an annual cycle? Also NCEP, ECMWF, COADS. Illustrates the overall uncertainty we have in the flux climatologies (since there is no "truth" here, this one is not beauty contest)
   
   
3. Flux model intercomparison. Using results from the Brunke et al. study, but maybe presented in a slightly different way. Compare obs, COARE 2.6, and the satellite bulk algorithms, show LH bias for the aggregate set of field deployments. Then show wind speed dependence of the LH biases for each algorithm if possible in single plot?
   
   
4. SST: pick one case in the tropics during the TRMM period, one of the ETL cases if possible with ancillary skin measurements (maybe JASMINE or NAURU'99?). Pick 3-4 week or so time series, compare in situ skin, bulk sst with NOAA operational SST product, AVHRR clear sky direct retrievals (Emery, Rossow), TMI values, and the Rossow and Clayson/Curry high res products. Probably 2 figs here.
   
   
5. Winds. Not so sure what to do here. Main issues seem to be resolution, and accuracy under highly disturbed conditions. Maybe the simplest thing to do in terms of a figure is to compare wind speeds (not vector wind fields) for Labrador Sea (KNORR) point (quikscat, ers, radarsat, ssmi, NCEP, ECMWF). This would highlight resolution, accuracy of the different products, and lead into discussion of attempts at merged wind products for higher res. I picked the Labrador Sea since Will Perrie has already been doing some of this for Labsea, and it is useful again to highlight some mid/high lat cases (to show that we care about something other than the tropics). Definitely open to better suggestions for a fig here
   
   
6. Any figs with Ta, qa, or further figs on satellite flux algorithms? Should we also show tropical case (in addition to labsea) where everyone can do a good job? Or just discussion?

0 comments

http://seaflux.gfdi.fsu.edu/article.php/20050912204214943