Comparison and Evaluation of Scatterometer (SCR) observed wind data with buoy wind data

1
Comparison and Evaluation of Scatterometer (SCR)
observed wind data with buoy wind data

• 
  Xinzhong Zhang
• 
  Remote Sensing
• 
  December 8th, 2009

2
Outline

• Abstract
• Introduction
• SCR observation issues
• Data Processing Results
• Future work

3
Abstract
4
Abstract
Seawinds scatterometer (SCR) wind data on
QuikSCAT are compared with buoy observed wind
data in this study. Based on this comparison, we
can evaluate the accuracy and reliability of SCR
wind data. Furthermore, factors like SST, ocean
current, rain effect etc., which could probably
affect the SCR wind data accuracy are briefly
reviewed.
5
Introduction
6
Introduction

• Why is this kind of comparison important?
• How does SCR wind observation work?
• How does buoy-based wind observation work?
• Potential factors to influence scatterometer wind
  data.
• Data description (SCR data Buoy wind data).

7
Why is this important?

• Increasing popularity of SCR wind data
• Errors INDEED exist within those SCR wind data
• Calibration is needed to improve SCR wind data
  accuracy

8
How does SCR wind observation work?

• Bragg scattering
• EMPIRICAL model function to relate backscatter
  cross section to Relative wind velocity

9
How does buoy-based wind observation work?
Anemometer

• Anemometer installed at the top
• Wind speed and direction are averaged over
  periods of 8 or 10 minutes

10
(No Transcript)
11
12 m
10 m
6 m
3 m
lt 2 m
12
Potential factors to influence SCR wind data.

• SST (Liu, W. T., 1984)
• Air-sea temperature difference
• Atmospheric density stratificaiton (Liu, 1984
  Wu, 1991)
• Underneath ocean currents (Kelly et al., 2001)
• Rain (Weissman et al.,2002)

13
Data description

• Seawinds data on QuikScat (ftp//ftp.ssmi.com/qsca
  t/qscat_wind_vectors/ )
• Original Orbit Wind Vector Data (swath data),
  which are not averaged spatially between
  ascending and descending swaths.
• Each orbit has one set of data, where those data
  are gridded into wind vector cells, 76(cross
  orbit) 1624 (along orbit)

• Buoy wind data (http//www.ndbc.noaa.gov/)
• Continuous wind data (10 minutes periods
  average)
• 10 minutes time interval
• Original wind data at the height of buoy
  anemometer, which need to be converted in to
  standardized height (10m) in order to be
  comparable to the satellite wind data (wind at
  10m height)

14
SCR observation issues
15
Satellite observed RELATIVE velocity (wind
relative to the sea surface)

• Ocean current effect to the QuikScat observed
  wind

(Xu, Y., R. B. Scott. 2008)
16

• Rain effect to the QuikScat observed wind

(Weissman, et al. 2001).
17

• SST effect (Left figure) and Atmospheric
  stability effect (Right figure) to the QScat
  observed wind

(Liu, W. T., 1984)
18
Data Processing Results
19
1. Select buoy station 44009, located right
outside of the Delaware Bay.
44009
Anemometer height 5 m above site elevation
Fortunate thing Its distance away from the coast
is gt25 km.
20
2. Choose time periods of interest. Here in
this study, the periods of Jan. 1st-10th, Apr.
1st-10th, Jul. 1st-10th, Oct. 1st-10th from 2005,
2006, 2007, 2008 years are chosen, so that both
available satellite data and buoy data could be
compared within those chosen periods.
21
3. Choose collocation time and distance ranges
for QuikScat observation and buoy observation
Buoy location
25km
QuikScat wind cells closest to the buoy locations
within 25 km and buoy observations closer to the
QuikScat winds within 30 minutes are chosen.
4. Convert buoy wind data to the equivalent
neutral wind speed at the standardized height
(10m), using the method of Liu and Tang (1996).
Satellite Swath
Wind at 10m
Wind at 5m
22
Results
For all collocated data available (193 data
points)
23
Results
For all collocated data available (193 data
points)
24
Results
Jan1-10 Apr1-10 Jul1-10 Oct1-10
R_wspeed 0.9121 0.9497 0.8875 0.9613
R_wdir 0.8356 0.8061 0.6233 0.5966
POOR!!
For 4 years averaged data
Should be due to the inappropriate processing
method
25
Future work
Find out
Jan1-10 Apr1-10 Jul1-10 Oct1-10
Wspeed_Qscat minus Wspeed_Buoy 0.0913 -0.3823 0.3551 -0.1603
Mean wind direction 272.3416 13.9619 139.093 273.320
Does the sign ( OR -) of the difference give any
information about the ocean current below the
surface wind?
Need Current data to verify!
26
Future work
Choose more interesting buoys and compare again.
27
Reference

• Dickinson, S., K.A. Kelly, M.J. Caruso, and M.J.
  McPhaden (2001) A note on comparisons between
  TAO buoy and NASA scatterometer wind vectors. J.
  Atmos. Oceanic Tech., 18, 799806.
• Freilich, M. H., and R. S. Dunbar (1999), The
  accuracy of the NSCAT 1 vector winds Comparisons
  with National Data Buoy Center buoys, J. Geophys.
  Res., 104(C5), 11,23111,246.
• Xu, Y., R. B. Scott. Subtleties in forcing eddy
  resolving ocean models with satellite wind data.
  Ocean Modelling 20 (2008) 240251
• Liu, W.T., Tang, W., 1996. Equivalent neutral
  wind, JPL Publication 96-17, Jet Propulsion
  Laboratory, Pasadena, 16 pp.
• LIU, W. T., 1984, The effects of the variations
  in sea surface temperature and atmospheric
  stability in the estimation of average wind speed
  by SEASAT-SASS, J. Phys. Oceanogr., 14, 392 401.
• Satheesan, K. Sarkar, A. Parekh, A. Kumar, M. R.
  Kuroda, Y., Comparison of wind data from QuikSCAT
  and buoys in the Indian Ocean. International
  Journal of Remote Sensing. 2007, VOL 28 Vol 10,
  pages 2375-2382
• Weissman, D. E., M. A. Bourassa, and J. Tongue,
  Effects of rain-rate and wind magnitude on
  SeaWinds scatterometer wind speed errors, J.
  Atmos. Oceanic Technol., submitted,2001.