Dear colleagues,

I want to reprocess 10 years of seaborne measurements of water-leaving radiance reflectance measurements made by abovewater radiometry and am writing for advice on where to obtain wind speed data.

Our protocol, based on the NASA Ocean Optics protocols and the simulations of Mobley (1999) and hence the wind/wave statistics of Cox and Munk (1954), requires input for wind speed at 10m above sea level in order to calculate a "wave-roughened Fresnel" reflection coefficient for sky radiance reflection at the air-sea interface. See equations (23) and (24) of:
http://aslo.org/lo/toc/vol_51/issue_2/1167.pdf

Over the years we have recorded this data locally from ship measurements, but I would like to now quality control all those wind speed measurements against a standardised dataset from a numerical weather model, preferably in data assimilated hindcast mode. Obviously there are differences in terms of spatial and temporal resolution.

Ideally I will send a list of about a few hundred sparsely distributed data points (lat/lon/date/UTC time) into an extraction tool and receive back a list of corresponding wind speeds. Any extra information, like surface atmospheric pressure, aerosol optical thickness, surface PAR, etc. could be a bonus but it is the windspeed that is most critical because it is used in the processing of the seaborne data.

If there are different data sources then spatial and temporal resolution will be the key criteria for selecting one data source over another. Temporal resolution should be 6-hourly or preferably better (1 hour?). Spatial resolution should be one degree or preferably better (10km?).

The data points are mainly located in the Southern North Sea and the English Channel but there are also points in the Mediterranean and a few in South American waters (French Guyana), so the data source needs to be global. Time period covered is 2001-2011.

The data extraction has to be very automatic. There are quite a few hundred locations to be processed.

I was initially thinking of a meteorological data source such as those used for ocean colour data processing (NCEP?), but there could well be other possibilities I have not considered.

Thanks in advance for any help you can provide.

Kevin Ruddick

The Ocean Color group at GSFC has the meteorological data used for its ocean color processing.  The surface winds are at 1 degree and every 6 h.  However, we have no available tool to easily grab a value from the files (I have some IDL code to roughly get values from climatology files, but it would require some work to adapt).  The NCEP source for this data is forward stream, with no hindcast available.  There is a MERRA data reanalysis that promises to have about .5 degree and 3 h resolution but I have not had time to look into using it yet.  They should have analysis from 1977 to near present.  So, that is where we stand at his time.  Perhaps, there is some better suited capability out there.  Otherwise, we can help in using our datasets.