SAF/OSI/CDOP/KNMI/TEC/MA/126
6. Data quality
As introduced in section 5.1, the accuracy should be better than 2 m/s in wind component
RMS with a bias of less than 0.5 m/s in wind speed.
Collocation result - speed (977034 wind vectors) Collocation result - direction (821351 wind vectors)
Model wind speed (m/s)
Collocation result - u (977140 wind vectors)
Model wind direction (deg)
Collocation result - v (977104 wind vectors)
Figure 2:
Contoured histograms of the
25-km ASCAT wind product from
Metop-A (top) and
Metop-B (bottom) vs. ECMWF model
forecast winds.
Model u component (m/s)
Histogram - speed
Model v component (m/s)
Histogram - direction
Collocation result - speed (966770 wind vectors)
Model wind speed (m/s)
Collocation result - u (966859 wind vectors)
Collocation result - direction (811903 wind vectors)
Model wind direction (deg)
Collocation result - v (966850 wind vectors)
Model u component (m/s)
Histogram - speed
Model v component (m/s)
Histogram - direction
Figure 2 shows two-dimensional histograms of the retrieved winds versus ECMWF 10m wind
background for the 25-km wind product, with the backscatter calibration (see section 3.2)
applied. The data for these plots are from consecutive Metop-A and Metop-B orbits from 1 and 2 November 2012 (28 orbits from both satellites).
The top left plots correspond to wind speed (bins of 0.5 m/s) and the top right plots to wind direction (bins of 2.5°). The latter are computed for ECMWF winds larger than 4 m/s. The bottom plots show the u and v wind component statistics (bins of 0.5 m/s).
OSI SAF/EARS ASCAT Wind Product User Manual version 1.13 May 2013 17
SAF/OSI/CDOP/KNMI/TEC/MA/126
From these results, it is clear that the spread in the distributions is small. The wind speed biases are quite small, ~0.20 m/s. We obtain wind component standard deviations of 1.44 in
u and 1.49 in v directions, both for Metop-A and Metop-B. Note that the data in the plots do not include WVCs that are quality controlled, i.e. have a too large inversion residual.
Collocation result - speed (6256 wind vectors) Collocation result - direction (5531 wind vectors)
Buoy wind speed (m/s)
Collocation result - u (6257 wind vectors)
Buoy wind direction (deg)
Collocation result - v (6257 wind vectors)
Figure 3:
Contoured histograms of the
25-km ASCAT wind product from
Metop-A (top) and
Metop-B (bottom) vs. in situ buoy
winds.
0
Buoy u component (m/s)
Histogram - speed
0
Buoy v component (m/s)
Histogram - direction
Collocation result - speed (6278 wind vectors) Collocation result - direction (5559 wind vectors)
Buoy wind speed (m/s)
Collocation result - u (6280 wind vectors)
Buoy wind direction (deg)
Collocation result - v (6280 wind vectors)
0
Buoy u component (m/s)
Histogram - speed
0
Buoy v component (m/s)
Histogram - direction
Figure 3 shows the results of collocations of ASCAT winds from Metop-A and Metop-B with buoy winds. These results cover the period of November 2012 to January 2013 (3 months).
For more details about the collocation method, please see [Ref-4] and [Ref-5]. We find a
wind speed bias of -0.02 m/s for Metop-A and 0.05 m/s for Metop-B. Standard deviations of u and v wind components are 1.72 m/s and 1.76 m/s for Metop-A and 1.78 m/s and 1.80 m/s for Metop-B, respectively. These figures are all well within the OSI SAF requirements.
More validation information can be found in [Ref-4] and [Ref-5].
OSI SAF/EARS ASCAT Wind Product User Manual version 1.13 May 2013 18
