[RADS] Wave Watch 3 model on Jason-1 data (experimental)
Remko Scharroo
remko at altimetrics.com
Thu May 1 17:52:06 CEST 2008
Dear RADS users,
** FIELDS 4001-4401: Wave Watch 3 model **
At the request of Doug Vandemark and Hui Feng at the University of New
Hampshire, I have now added FIVE new data fields to RADS data base.
They are:
field 4001 (wave_m0) Variance of the wave height [m2]
field 4101 (wave_m1) First moment of the wave height [m2/s]
field 4201 (wave_m2) Variance of the wave velocity [m2/s2]
field 4301 (wave_m4) Variance of the wave slope [radian2]
field 4401 (wave_shs) Height of wave swell [m]
Each of these variable comes from the Wave Watch 3 model, Version
2.22, as computed at UNH/OPAL and kindly made available to the RADS
project by Doug and Hui. The model has a resolution of 1ºx1ºx6h and is
available for all latitudes between 70ºN and 70ºS and for the years
2000-2005.
The grids have a changing mask, eliminating land and ice cover. I
interpolate the grids linearly in space and time, each of the 8
closest grid nodes get a weight based on the relative distance from
the interpolation point to the node. The total weight can be 1.0
maximum, but can be lower when data are masked out. The interpolation
point gets a value when the weight is 0.5 or greater, thus allowing
some continuation close to the coasts. Doug, however, wants me to
stress that the model should be trusted only on the open ocean.
The variance of wave height has some similarities to the significant
wave height. The relationship is roughly:
SWH = 4 sqrt(m0). However, these quantities are based on different
definitions of wave height. SWH is the average height of the highest
1/3 of the waves, while m0 is the variance of all waves throughout the
spectrum. Since SWH only focusses on the small selection of the wave
spectrum, a skew in the wave height distribution will change that
relationship between SWH and m0.
Attached is a quick example of the comparison: altimeter SWH in red,
4*sqrt(m0) in green.
The model data have currently only been added to Jason-1 cycles 1-9
for testing. If there are no clear errors in the implementation and
there is general interest in this model (so PLEASE CHIME IN!), then
I'll add it to all altimeter data of the 2000-2005 period.
Have a nice May Day!
Remko
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ww3.png
Type: image/png
Size: 139439 bytes
Desc: not available
Url : http://www.deos.tudelft.nl/pipermail/rads/attachments/20080501/200dbedd/attachment.png
-------------- next part --------------
More information about the RADS
mailing list