SIW TAC Scientific Calibration report
Applicable documents, AD-1 : MYO-SIWTAC-Cal/Val Plan, V1.0, 21. June 2010
Surface Wind Climatology
I.1
Overall description of calibration task
AD-1: The objective of the calibration is to perform verification of the SIW TAC products before release
of the products into the V1 portfolio.

For existing products which are introduced into the TAC, the aim of the calibration is: To
verify that the production gives the expected results and that this is disseminated
correctly.

For new developed products a systematic analysis of the quality is necessary. Calibration
is performed by comparison with high quality reference data set in a limited time period.
The surface wind climatology is a new product provided by IFREMER as V2 product. It is available
over global ocean.. The products are provided every month.
The calibration will be:
1) Verify that the products are gridded correctly and CF-1.4 compliant
2) Intercomparison with in-situ monthly wind estimates.
I.2
Description of calibrated system
The climatology refers to time series of monthly averaged wind variables calculated over global
oceans. It is estimated from daily global wind fields calculated from retrievals derived from ASCAT
scatterometer onboard METOP-A satellite. It consists of six variables including monthly averaged
wind speed, zonal and meridional wind components, wind stress amplitude and the associated
components. They are calculated as arithmetic means of ASCAT daily wind analyses. The gridded
daily wind and wind stress fields have been estimated over global oceans from Metop/ASCAT
retrievals using objective method. The analyses use standard products ASCAT L2b during the period
April 2007 to present. ASCAT retrievals occurring during April 21, 2007 to November 20, 2008 are
calculated as real winds. They are converted to equivalent neutral winds using bulk parametrisation
Coare3.0 (Fairall et al, 2003). Coare3.0 parameterization is also used for wind stress estimations.
They are calculated over ASCAT swaths from observations. The resulting fields are estimated as
equivalent neutral-stability 10-m daily winds, and have spatial resolutions of 0.25° in longitude and
latitude. The objective method aims to provide daily-averaged gridded wind speed, zonal component,
meridional component, wind stress and the corresponding components at global scale. The error
associated to each parameter, related to the sampling impact and wind space and time variability, is
provided too. More details about data, objective method, computation algorithm may be found in
(Bentamy et al, 2011). For monthly calculation purpose, only valid daily data available within each
month of the period 2007 trough 2011 are used. The monthly winds are estimated at each grid point
(0.25°0.25°) from at least 25 daily values available at the same grid point. The associated root mean
square (rms) values are also calculated at each grid point and used as quality control variable. The
resulting final files are provided as MyOcean V2 products in CF-1.4 compliant NetCDF format.
The NetCDF-files are finally transmitted to the Dissemination Unit at Ifremer. A verification of the data
content and quality has been performed. The netCDF file is checked against the CF-Convention
compliance checker for NetCDF format (http://cf-pcmdi.llnl.gov/conformance/compliance-checker/)
I.3
Calibration results
Output from the compliant check of a sample NetCDF file:
File name: 2011051612_1mm-ifremer-L4-EWSB-wind_gridded-GLO-20110902161227NRT-02.0.nc
Output
of
CF-Checker
follows...
CHECKING
NetCDF
FILE:
/tmp/6351.nc
=====================
Using
CF
Checker
Version
2.0.2
Using
Standard
Name
Table
Version
16
(2010-10-11T12:16:51Z)
Using
Area
Type
Table
Version
1
(5
December
2008)
-----------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
----------------------------------Checking
------------------
variable:
variable:
variable:
wind_speed
sampling_length
wind_stress
variable:
surface_downward_eastward_stress
variable:
eastward_wind_rms
variable:
surface_downward_northward_stress
variable:
northward_wind
variable:
wind_speed_rms
variable:
variable:
variable:
northward_wind_rms
land_ice_mask
surface_downward_northward_stress_rms
variable:
variable:
longitude
surface_downward_eastward_stress_rms
variable:
variable:
depth
time
WARNING (4.4.1): Use of the calendar and/or month_lengths attributes is recommended for time
coordinate
variables
-----------------Checking
variable:
latitude
----------------------------------Checking
variable:
wind_stress_rms
----------------------------------Checking
variable:
eastward_wind
-----------------ERRORS
detected:
0
WARNINGS
given:
1
INFORMATION messages: 0
To meet the calibration definition of V2 products (see above), the quality of the resulting climatology is
performed trough comprehensive comparisons with monthly winds from moored buoy data. The latter
are derived from various buoy networks: NDBC/NOAA (Atlantic, Pacific oceans), UK Met Office and
Météo-France (Atlantic and Mediterranean Sea), TAO (Tropical Pacific), PIRATA (Tropical Atlantic),
and RAMA (Indian Ocean). More than 170 buoy raw data are collected, investigated, and collocated in
space and time with satellite estimates. Comparisons between monthly averaged data from buoys and
from satellites are quite challenging. Indeed the two sources do not measure surface based on same
space and time resolution. Figure 1 shows examples of wind speed times series from buoy (in green)
and from the satellite climatology (in red) at three NDBC buoy locations : 25.90°N – 89.67°W (a: Gulf
of Mexico), 43.19°N – 69.16°W (b: Atlantic), 37.76°N – 122.83°W (c: Pacific). Both time series track
each other well. The main temporal features exhibit by boy estimates are clearly depicted by satellite
winds. To assess the quality of surface wind climatology, all collocated satellite and buoy data,
including nearshore and offshore locations, are used. Figure 2, 3, 4, and 5 illustrate the comparisons
through scatterplots between buoy and satellite monthly wind speeds, wind directions, wind stress
zonal components, and wind stress meridional components, respectively. The statistical parameters
bias, root mean square (rms) difference (buoy – satellite in this order) and scalar correlation (Cor) and
vector correlation (for wind direction) are calculated. Their values are provided in each figure.
Figure 1 : Time series of monthly wind speed from buoy measurements and from satellite climatology
In general, the four variables derived from buoys and from satellite climatology are in good agreement.
No significant systematic biases are depicted. More specifically, for wind speed, the bias is quite
smalland the rms difference is about 0.80m/s. The associated scalar correlation is almost 0.90. For
wind direction, a small bias is found and mainly related to the comparisons performed at nearshore
locations and for low wind speed conditions.
Figure 2 : Comparison between collocated monthly wind speeds from buoys and from satellite climatoalogy
during the period : 2007 – 2010.
Figure 3 : Comparison between collocated monthly wind directions from buoys and from satellite
climatoalogy during the period : 2007 – 2010.
Figure 4: Comparison between collocated monthly wind stress zonal components from buoys and from
satellite climatoalogy during the period : 2007 – 2010.
Figure 5 : Comparison between collocated monthly wind stress meridional components from buoys and
from satellite climatoalogy during the period : 2007 – 2010.
I.4
Calibration technical summary
Production Centre:
SIW-IFREMER-BREST-FR
Calibration task:
The product is checked for compliance with NetCDF/CF-1.4 convention, and is compared to in-situ
surface wind estimates.
Calibration summary:
The NetCDF product complies technically with the CF-1.4 convention.
The wind climatology is compared with the product is compared with wind speed, directions, wind
stress components derived from NDBC buoy measurements. The agreements are very good.
(success/problems, half page max)