remote sensing
Article
Hurricane Wind Speed Estimation Using WindSat 6
and 10 GHz Brightness Temperatures
Lei Zhang 1,2 , Xiao-bin Yin 2,3,4, *, Han-qing Shi 1 and Zhen-zhan Wang 2,3
1
2
3
4
*
Institute of Meteorology and Oceanography, PLA University of Science and Technology,
Nanjing 211101, China; zlei_best@hotmail.com (L.Z.); mask1000@126.com (H.S.)
Key Laboratory of Microwave Remote Sensing, Chinese Academy of Sciences, Beijing 100190, China;
wangzhenzhan@mirslab.cn
National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China
Beijing Piesat Information Technology Co., Ltd., Beijing 100190, China
Correspondence: yinxiaobin@mirslab.cn; Tel.: +86-10-6258-2844
Academic Editors: Xiaofeng Li and Prasad S. Thenkabail
Received: 4 August 2016; Accepted: 23 August 2016; Published: 31 August 2016
Abstract: The realistic and accurate estimation of hurricane intensity is highly desired in many
scientific and operational applications. With the advance of passive microwave polarimetry,
an alternative opportunity for retrieving wind speed in hurricanes has become available. A wind
speed retrieval algorithm for wind speeds above 20 m/s in hurricanes has been developed by using
the 6.8 and 10.7 GHz vertically and horizontally polarized brightness temperatures of WindSat.
The WindSat measurements for 15 category 4 and category 5 hurricanes from 2003 to 2010 and the
corresponding H*wind analysis data are used to develop and validate the retrieval model. In addition,
the retrieved wind speeds are also compared to the Remote Sensing Systems (RSS) global all-weather
product and stepped-frequency microwave radiometer (SFMR) measurements. The statistical results
show that the mean bias and the overall root-mean-square (RMS) difference of the retrieved wind
speeds with respect to the H*wind analysis data are 0.04 and 2.75 m/s, respectively, which provides
an encouraging result for retrieving hurricane wind speeds over the ocean surface. The retrieved
wind speeds show good agreement with the SFMR measurements. Two case studies demonstrate
that the mean bias and RMS difference are 0.79 m/s and 1.79 m/s for hurricane Rita-1 and 0.63 m/s
and 2.38 m/s for hurricane Rita-2, respectively. In general, the wind speed retrieval accuracy of the
new model in hurricanes ranges from 2.0 m/s in light rain to 3.9 m/s in heavy rain.
Keywords: polarimetric microwave radiometer; sea surface wind retrieval; WindSat; H*wind
analysis data
1. Introduction
Passive microwave remote sensing is an important tool for studying atmospheric and
oceanographic processes and can provide both daytime and nighttime observations of geophysical
parameters [1], such as atmospheric water vapor, cloud liquid water, rain rate, sea surface temperature,
sea surface wind, and sea ice. This tool can provide scientists and forecasters with vital information for
understanding and studying global weather and climate change. The first Special Sensor Microwave
Imager (SSM/I) aboard the Defense Meteorological Satellite Program (DMSP) was launched in
July 1987 [2], making it possible to obtain the global ocean surface wind from a spaceborne passive
microwave instrument [3]. Since then, a successive dataset of wind speed has been obtained from the
passive microwave radiometry [3–5].
Algorithms [3,6–9] have been developed for passive microwave radiometry that are able to
retrieve the wind speeds of low to moderate winds in no-rain areas with a good degree of accuracy.
Remote Sens. 2016, 8, 721; doi:10.3390/rs8090721
www.mdpi.com/journal/remotesensing
Remote Sens. 2016, 8, 721
2 of 20
However, these algorithms seem to break down completely when the scenes are under severe weather
conditions, including both high winds and precipitation. It is difficult to measure wind speeds in these
conditions because the physics of ocean surface remote sensing is still poorly understood under these
weather conditions. Intense rain not only influences the atmospheric attenuation but also changes the
sea surface dielectric properties in a complicated manner [10,11]. Thus, it is very difficult to accurately
model brightness temperatures in these situations. Fortunately, the brightness temperature (TB)
acquired at the lower frequencies, such as the L-band and C-band, are far from saturation, which is
the physical reason that researchers can “see” the ocean surface and derive its properties [11,12].
A new capability for ocean surface wind speed retrieval is performed using the L-band brightness
temperatures provided by the Soil Moisture and Ocean Salinity (SMOS) mission [11]. The study [11]
shows that the L-band ocean emissivity dependence with wind speed appears to be less sensitive to
roughness and foam changes than at the higher C-band microwave frequencies. A physical method is
developed to retrieve sea surface temperature and wind speed for hurricane predictions in [13]. In [13],
the effects of precipitation emission and scattering on the measurements are both taken into account.
However, the used ocean microwave emissivity model is performed in clear sky conditions and at
low wind speeds (smaller than 25 m/s). Hence, the performance of the model needs to be further
tested at high wind speeds and intensely rainy conditions. In addition, the C-band H-polarization TBs
provided by AMSR-E and AMSR2 are used to retrieve wind speed in hurricanes using the small slope
approximation method and an ocean surface roughness spectrum [14]. In [14], the retrieved wind
speeds are vulnerable to rain because the model does not account for the effects of absorption and
scattering caused by rain.
It is very difficult to accurately model brightness temperatures in rain because of the high
variability in rainy atmospheres. Thus, the statistical algorithm seems to be a good candidate for
retrieval of wind speeds in this situation. For accurate radiometer retrievals of wind speeds under
rainy conditions, channel combination methods at different frequencies have been used to retrieve
wind speeds [10,12,15,16]. The core of these algorithms makes it possible to reduce the signal coming
from rain without considerably reducing the wind speed signal. Such a technique has been employed
successfully for wind speed retrieval with the airborne microwave polarimetry. In addition, an artificial
neural network technique has been used to derive the relationship between the AMSR2 TBs and sea
surface wind speeds in hurricanes [17–19]. However, the retrieved results are mainly compared with
the in situ surface wind speed observations below 30 m/s in [17–19]. They are not validated using the
high wind measurements.
In this study, a new model for retrieving high wind speeds in hurricanes using WindSat C-band
and X-band vertically and horizontally polarized brightness temperatures is proposed. The core of
the new model is also based on the channel combination at different frequencies. Referring to [15],
we define two new parameters, W6H and W6V, which represent an increment at 6H and 6V due to
ocean wind, respectively. Finally, a wind speed retrieval algorithm for wind speeds above 20 m/s in
hurricanes has been developed by using the two new parameters. The WindSat measurements for
15 category 4 and category 5 hurricanes from 2003 to 2010 and the corresponding H*wind analysis
data are used to develop and validate the retrieval model.
This paper is organized as follows. In Section 2, the datasets, including WindSat TBs, H*wind
analysis data, SFMR measurements, RSS global all-weather product and rain rate, are introduced.
The methodology is presented in Section 3. Section 4 gives detailed results. The last section presents
the conclusions.
2. Datasets
In this paper, WindSat TB measurements over hurricanes, and H*wind analysis data are used to
develop and test a new wind speed retrieval algorithm for inside hurricanes. The SFMR along-track
wind speed measurements and the RSS global all-weather wind speed product are compared with the
Remote Sens. 2016, 8, 721
3 of 20
retrieved wind speeds. In addition, the rain effect is also evaluated by using the RSS rain rate product.
The selected datasets are described in the following sections.
2.1. Wind Sat Brightness Temperature
WindSat is a conically scanning passive microwave radiometer aboard the U.S. Department
of Defense Coriolis satellite, which orbits the Earth in an 840-km circular sun-synchronous orbit.
Its primary objective is to demonstrate the capability of a polarimetric microwave radiometer to
measure the ocean surface wind vector from space. WindSat operates in 5 channels: 6.8, 10.7, 18.7, 23.8,
and 37 GHz. All are fully polarimetric except for the 6.8 and 23.8 GHz channels, which feature
dual polarization (vertical and horizontal). The integrated field of view (IFOV) is 40 × 60 km,
25 × 38 km, 16 × 27 km, 12 × 20 km, and 8 × 13 km for the 6.8, 10.7, 18.7, 23.8, and 37 GHz
channels, respectively. WindSat takes TB observations during both the forward-looking and the
aft-looking scans. The forward-looking swath is approximately 1000 km and the aft-looking swath is
approximately 400 km. During each scan, every channel is of a different location and resolution. Hence,
the TB measurements must be resampled and averaged to a common size for wind vector retrieval.
The selected TB measurements have been averaged to the 6.8 GHz footprint size with the WindSat
ground data processing version 2.0 algorithm. A detailed description of the WindSat instrument can
be found in [20]. In this paper, we only use the vertically and horizontally polarized measurements for
wind speed retrieval.
2.2. H*wind Products
Validation of hurricane surface wind fields is inherently challenging because there is no realizable
“surface truth” data. Fortunately, a relatively reliable and reasonable estimation of tropical cyclone
(TC) wind fields has been developed through the H*wind analysis system. The wind fields produced
by this system have been extensively used to study the wind speed retrieval algorithms under severe
weather conditions [10,11,15,21–24]. H*wind is a software tool that provides an objective analysis of
the 1-min sustained-wind field at a height of 10 m above the sea surface and was developed by the
Hurricane Research Division (HRD) of the Atlantic Oceanographic and Meteorological Laboratory [25].
The H*wind analysis system assimilates all available surface observations, such as buoys, ships,
C-MAN platforms, and surface airways, and all available aircraft and remote sensing data sources,
which typically includes the SFMR, QuikSCAT, SSM/I, Geostationary Operational Environmental
Satellite (GOES) low-level cloud drift winds, and Global Positioning System (GPS) dropwindsondes
from aircraft. Typical wind speed errors in an H*wind analysis are estimated to be 10% to 20% [26].
The wind speed errors in this analysis will vary depending on the quantity and quality of data that are
available, as well as on the degree of quality control employed by the analyst. Although each wind
speed analysis produced by this system may be inaccurate, the ensemble average over a large number
of collocations can minimize the error in the overall analysis.
A hurricane is defined as storm with wind speeds exceeding 65 knots (33.5 m/s) [15].
The hurricanes selected for this paper are Fabian and Isabel in 2003; Frances and Ivan in 2004; Dennis,
Katrina, Rita, and Wilma in 2005; Helene in 2006; Felix in 2007; Bertha and Ike in 2008; Bill and
Ida in 2009; and Igor in 2010. The hurricane best track analysis and the corresponding intensity are
depicted in Figure 1. These data can be obtained from the Automated Tropical Cyclone Forecast
system [27], with products available at the National Oceanic Atmospheric Administration (NOAA)
National Hurricane Center (NHC).
Remote Sens. 2016, 8, 721
Remote Sens. 2016, 8, 721
4 of 20
4 of 20
o
(m/s)
60 N
(m/s)
70
o
48 N
60
o
50
o
40
o
30
70
o
60 N
60
(m/s)
(m/s)
70
o
60 N
60
o
70
o
60 N
60
48 N
36 N
24 N
12 N
0
o
o
o
45 N
50
o
40
o
30
o
(m/s)
40
o
24 N
15 N
20
o
0 o
o
o
o
o
90 W 75 W 60 W 45 W 30 W
72 W 60 W 48 W 36 W 24 W
50
o
36 N
30 N
20
o
o
o
(m/s)
o
30
o
12 N
0
o
o
o
o
48 N
70
48 N
70
o
60
42 N
o
60
o
60
o
50
36 N
o
50
o
50
o
40
30 N
o
40
o
30
24 N
30 N
24 N
18 N
30
o
o
o
o
o
o
20
18 N
o
90 W84 W78 W72 W66 W60 W
70
60
o
40 N
50
o
30 N
40
o
o
o
48 N
70
o
60
o
50
40 N
32 N
40
30
o
30
o
16 N
o
10 N
o
o
o
o
o
20
o
60 W 50 W 40 W 30 W 20 W
o
o
o
o
88 W 80 W 72 W 64 W 56 W
(m/s)
o
54 N
20
70
45 N
o
60
o
50
70
o
40
60
30 N
o
18 N
o
o
o
o
o
72 W 63 W 54 W 45 W 36 W
40
20
12 N
20
o
(m/s)
70
o
60
40 N
50
30 N
40
o
20 N
30
o
o
o
o
o
20
o
10 N
o
o
60 N
o
o
o
o
o
20
90 W 80 W 70 W 60 W 50 W 40 W
(m/s)
70
o
50 N
(m/s)
70
o
50 N
60
40 N
o
60
o
50
o
30 N
50
o
40
20 N
o
40
o
30
o
20
40 N
30 N
20 N
10 N o
o
o
o
o
o
70 W 60 W 50 W 40 W 30 W 20 W
o
10 N
o
0 o
o
o
o
o
o
100 W 90 W 80 W 70 W 60 W 50 W
30
20
(m/s)
o
60 N
o
70
60
50
40 N
o
40
30 N
o
30
o
o
24 N
18 N
o
o
30
50
o
27 N
o
50 N
o
36 N
o
(m/s)
o
36 N
o
90 W 75 W 60 W 45 W 30 W
50 N
96 W 88 W 80 W 72 W 64 W
(m/s)
o
0
o
o
40
20
o
24 N
o
20 N
32 N
96 W90 W84 W78 W72 W66 W
(m/s)
o
50 N
o
30
15 N
o
16 N
o
o
12 N
40 N
24 N
o
40
o
(m/s)
o
70
36 N
50
o
30 N
84 W72 W60 W48 W36 W
o
42 N
45 N
20
o
o
o
30
o
o
o
o
o
o
96 W 90 W 84 W 78 W 72 W
20
o
30
o
20
20 N
10 N o
o
o
o
o
o
70 W60 W50 W40 W30 W20 W
Figure 1. Best
Best track
track positions
positions and
and the
thecorresponding
corresponding intensity
intensity of
of the
the selected
selected hurricanes
hurricanes for
for (a)
(a) Fabian,
Fabian,
27 August–9 September
September 2003;
2003; (b)
(b) Isabel,
Isabel, 6–19
6–19 September
September 2003;
2003; (c)
(c) Frances,
Frances, 25
25 August–9
August–9 September,
September,
2004; (d)
4–17
July
2005;
(f) (f)
Katrina,
23-30
August
2005;
(g)
(d) Ivan,
Ivan, 2–24
2–24September
September2004;
2004;(e)(e)Dennis,
Dennis,
4–17
July
2005;
Katrina,
23-30
August
2005;
Rita,
18–26
September
2005;
(h)
Wilma,
15–26
October
2005;
(i)
Helene,
12-26
September
2006;
(j)
Felix,
(g) Rita, 18–26 September 2005; (h) Wilma, 15–26 October 2005; (i) Helene, 12-26 September 2006;
31
August–6
September
2007; (k)2007;
Bertha,
July
2008;
(l)2008;
Ike, 1–15
September
2008; (m)
Bill,
15–25
(j) Felix,
31 August–6
September
(k) 3–21
Bertha,
3–21
July
(l) Ike,
1–15 September
2008;
(m)
Bill,
August,
2009; 2009;
(n) Ida,
2009; 2009;
and (o)
Igor,
2010. 2010.
15–25
August
(n)4–11
Ida, November
4–11 November
and
(o) 8–22
Igor, September
8–22 September
2.3.
Radiometer (SFMR)
(SFMR) Measurements
Measurements
2.3. Airborne
Airborne Stepped-Frequency
Stepped-Frequency Microwave
Microwave Radiometer
An
sensing instrument
An SFMR
SFMR is
is an
an airborne
airborne remote
remote sensing
instrument designed
designed for
for retrieving
retrieving surface
surface wind
wind speed
speed
and
rainfall
in
TCs
[28].
The
first
operational
SFMR
was
flown
on
the
WP-3D
aircraft
in
and rainfall in TCs [28]. The first operational SFMR was flown on the WP-3D aircraft in 2004
2004 [29].
[29].
An SFMR
SFMR observes
observesthe
thenadir
nadirbrightness
brightness
temperatures
6 C-band
frequencies,
An
temperatures
at 6atC-band
frequencies,
i.e., i.e.,
4.55,4.55,
5.06,5.06,
5.64, 5.64,
6.34,
6.34,
6.96,7.22
andGHz.
7.22 GHz.
The
TB measurements
are processed
with a 10-second
to
6.96, and
The TB
measurements
are processed
with a 10-second
runningrunning
mean tomean
retrieve
retrieve
oceanwind
surface
wind
speed
using a geophysical
model
function
thatsurface
relatesemissivity
the surface
ocean surface
speed
using
a geophysical
model function
that
relates the
to
emissivity
to
wind
speed.
Global
Positioning
System
dropwindsonde
measurements
have
been
used
wind speed. Global Positioning System dropwindsonde measurements have been used to validate the
to
validate
SFMR
wind speed
and the
RMS error is approximately
4 m/s for
measured
SFMR
windthe
speed
estimates,
and estimates,
the RMS error
is approximately
4 m/s for measured
wind
speeds
wind
speeds
to 70
[29]. Astudy
previous
that the
SFMR
wind
speed
ranging
fromranging
10 to 70from
m/s10
[29].
A m/s
previous
[30] study
shows[30]
thatshows
the SFMR
wind
speed
in heavy
in
heavy
precipitation
has
been
further
improved
by
using
the
new
relationship
between
microwave
precipitation has been further improved by using the new relationship between microwave absorption
absorption
andThe
rainadvantage
rate. The advantage
SFMR
is that
it can potentially
retrieve
oceanwind
surface
wind
and rain rate.
of SFMR isofthat
it can
potentially
retrieve ocean
surface
speed
at
speed
at ahigh
relatively
spatial
(1.5 itkm).
However,
it point
can measurements
only provide along
point
a relatively
spatial high
resolution
(1.5resolution
km). However,
can only
provide
measurements
along
the aircraft flight track.
the aircraft flight
track.
2.4. Other Sources
Using WindSat TB measurements, a global all-weather wind speed retrieval algorithm has been
established by RSS [10]. It is a semi-statistical algorithm that can be applied in all rainy atmospheres
over the whole Earth. The estimated RMS accuracy of this algorithm ranges from approximately 2.0
m/s in light rain to approximately 5.0 m/s in heavy rain, if C-band channels and higher frequencies
Remote Sens. 2016, 8, 721
5 of 20
over the whole Earth. The estimated RMS accuracy of this algorithm ranges from approximately
2.0 m/s in light rain to approximately 5.0 m/s in heavy rain, if C-band channels and higher frequencies
are available [10]. It is a good candidate for comparison with our retrieved wind speed results using
WindSat TB measurements because the sufficient matchups can be obtained from them. We obtain the
latest RSS v7.0.1 all-weather wind speed products from the RSS website (http://www.remss.com/).
To analyze the effect of rain, the RSS WindSat v7.0.1 rain rate product is used to directly study
the wind speed retrieval errors. The other independent rain rate products are not selected because
the time difference may result in a spatial mismatch [31]. For example, a TC with a forward motion
of 10 m/s could travel 24 km in 40 min, comparable to the mean spatial resolution of the selected
TB. In addition, the distribution of rainfall inside TCs is complex and changeable. To some extent,
the mismatch could introduce some errors. We obtain the rain rate product from the RSS website
(http://www.remss.com/), and the product was developed using the Unified Microwave Ocean
Retrieval Algorithm (UMORA) [32].
2.5. Matchup
For training and testing the wind speed retrieval algorithm in TCs, we collected TB measurements
over the H*wind analysis fields. For WindSat, we collected data from 15 hurricanes during the period
from 2003 to 2010. The time of the H*wind analysis data and the average time of the WindSat overpass
are within 3 h. No temporal interpolation is performed as the time intervals are too large in most cases.
Within the 3-h time window between the products, the hurricane could have moved over a relatively
large distance in some cases, which could introduce significant error. Thus, we shift the H*wind
analysis data for each TC based on the “best track” fixes from NHC so that the eye of the H*wind
analysis data coincides with the eye of the WindSat measurements. In addition, the H*wind analysis
fields are performed at a high resolution, which is approximately 5 km [25].
We resample the H*wind analysis data to make them comparable to the low-resolution (6.9 GHz)
WindSat TB measurements. Using a nearest neighbor interpolation, all H*wind data points within
a radius of 30 km (around the size of the WindSat 6.8 GHz IFOV) of the WindSat TB locations are
weighted, inversely proportional to distance, and then averaged:
n
∑ H ∗ wind (i) × weight (i)
WS =
i=1
(1)
n
∑ weight (i)
i=1
The function of weight is
weight (i) = exp
−dis (i) × dis (i)
4 × 30
(2)
where the “dis” is the distance (in kilometers) from an H*wind point to a WindSat grid point. The form
of weighting function is selected because it is based on the fact that the power that is received by the
radiometer falls off roughly exponentially with distance and that wind speed is roughly proportional
to the brightness temperature (power).
A previous study has shown that the radiometer TB is directly sensitive to surface roughness
rather than wind speed [10]. The H*wind analysis reports peak wind speeds for 1-min sustained
winds. Thus, we need to convert from 1-min sustained H*wind data to 10 min sustained satellite wind
speeds because the surface winds for producing the surface roughness of the satellite measurements
need to be sustained a longer time, which is at least 10 min. A derived scale factor of 0.88 is used here
and by the U.S. Navy [33].
Remote Sens. 2016, 8, 721
6 of 20
Table 1 lists the statistics of the selected TCs in the datasets, which includes the TC name, year,
maximum wind speed, the latitudinal and longitudinal shifts of the TC eye, and the population of the
selected wind speed above 20 m/s. Finally, the collocations contain 26,242 matchups for the WindSat.
Table 1. Statistics of matchups between WindSat observations and H*wind analysis data for each
selected TC.
Fabian
Isabel
Frances
Ivan
Dennis
Katrina
Rita
Wilma
Helene
Felix
Bertha
Ike
Bill
Ida
Igor
Year
Max Winds
(m/s)
2003
2003
2004
2004
2005
2005
2005
2005
2006
2007
2008
2008
2009
2009
2010
60
73
58
70
65
75
78
80
53
73
53
63
58
45
68
WindSat-H*wind
Population
(ws > 20m/s)
Longitude
Shift (◦ )
Latitude
Shift (◦ )
1451
3633
2390
4012
1021
798
2673
932
1409
555
1275
2199
376
548
2970
−0.054
−0.133
−0.018
−0.318
0.129
0.233
0.362
−0.169
0.042
0.280
−0.135
−0.006
−0.225
0.028
−0.047
−0.287
−0.058
0.166
−0.264
0.033
−0.050
−0.028
−0.223
−0.050
−0.132
−0.053
−0.054
−0.346
−0.077
−0.033
3. Methodology
3.1. Polarimetric Microwave Radiation
The Stokes parameters are a set of values that characterize the polarization state of the
electromagnetic radiation. Four parameters, I, Q, U, and V, were defined by George Gabriel Stokes
in 1852. Since passive microwave radiometers usually provide vertically and horizontally polarized
brightness temperatures (TV and TH ) to observe the Earth, an alternate Stokes vector can be represented
as TV , TH , T3 , and T4 :
E 
 D


|EV |2
TV
E 
 D
 T 
2


 H 
E
|
|
H

TB = 
(3)
 = c×


 T3 
∗
 2Re hEV EH
i 
T4
2Im hE E∗ i
V H
where TV and TH describe the brightness temperatures of vertical and horizontal polarizations,
respectively. T3 and T4 are the cross-correlation terms between these two orthogonal polarizations.
In addition, c is a constant relating the brightness temperature to the electric energy density [34,35].
In recent years, various geophysical parameters, such as the wind vector, sea surface temperature,
atmospheric water vapor, cloud liquid water, rain rate, and soil moisture, have been estimated by
using the polarimetric brightness temperatures [8,9,36,37]. The brightness temperatures measured by
the spaceborne polarimetric radiometer are the sum of the upwelling atmospheric radiation (TBU ),
the reflected cosmic background radiation (TBC ) and the downwelling atmospheric radiation (TBD ),
and the emission of the ocean surface. The atmosphere attenuates the emission of the ocean surface
and the reflected downwelling radiation. Thus, the four Stokes parameters received by a spaceborne
polarimetric radiometer at each frequency can be expressed as follows [34]:
Tv = TBU + τ [ev TS + rv (Ωv TBD + τTBC )]
(4)
Remote Sens. 2016, 8, 721
Remote Sens. 2016, 8, 721
7 of 20
Th = TBU + τ e h TS + rh ( Ω h TBD + τTBC ) 
7 of 20
Th = T
TBU=+τeτ [ehTTS−+(rTh (Ω+h TτBD
T +) τTBC )]
(5)
(6)
3
3

S
BD
BC

(5)
T3 = τe3 [TS − (TBD + τTBC )]
(6)
(7)
(7)
T4 = τe4 [TS − (TBD + τTBC )]
where ep refers to the ocean surface emissivity at polarization p (p = v, h), and the corresponding
where ep refers to the ocean surface emissivity at polarization p (p = v, h), and the corresponding ocean
ocean surface reflectivity is rp = 1 − ep. TS is the sea surface temperature, and τ is the transmissivity
surface reflectivity is rp = 1 − ep . TS is the sea surface temperature, and τ is the transmissivity of the
Ω p (p
of
the atmosphere.
v, h)isterm
is a correction
to account
for nonspecular
reflection
atmosphere.
The ΩThe
h) =term
a correction
factor factor
to account
for nonspecular
reflection
of the
p (p = v,
TBD
the rough
oceanocean
surface
[38]. [38].
Figure
2 describes
the the
brightness
temperatures
received
by
the rough
surface
Figure
2 describes
brightness
temperatures
received
of
thefrom
TBD from
a radiometer.
by
a radiometer.
T4 = τe 4 TS − ( TBD + τTBC ) 
Figure 2. The brightness temperatures received by a radiometer.
Figure 2. The brightness temperatures received by a radiometer.
3.2. Channel Selection
3.2. Channel Selection
Under clear weather conditions, it is common to select higher frequencies such as 36 GHz rather
Under clear weather conditions, it is common to select higher frequencies such as 36 GHz
than lower frequencies for retrieving the low to moderate ocean surface winds because the sensitivity
rather than lower frequencies for retrieving the low to moderate ocean surface winds because the
to ocean surface wind at higher frequencies is better than that at lower frequencies [15,39]. However,
sensitivity to ocean surface wind at higher frequencies is better than that at lower frequencies [15,39].
the higher frequencies are almost invalid for retrieval of high winds in the severe weather conditions
However, the higher frequencies are almost invalid for retrieval of high winds in the severe weather
because the atmospheric opacity for higher frequencies exceeds that for lower frequencies. Figure 3
conditions because the atmospheric opacity for higher frequencies exceeds that for lower frequencies.
shows the WindSat brightness temperatures for hurricane Isabel on 10 September 2003, along with
Figure 3 shows the WindSat brightness temperatures for hurricane Isabel on 10 September 2003, along
the best track (piecewise red curve) analysis from the NHC. The top three panels of Figure 3 show
with the best track (piecewise red curve) analysis from the NHC. The top three panels of Figure 3
the results of horizontally polarized brightness temperature for the 10.7, 18.7, and 37 GHz channels,
show the results of horizontally polarized brightness temperature for the 10.7, 18.7, and 37 GHz
respectively. Note that we use a different color scale for the 10.7, 18.7, and 37 GHz images to highlight
channels, respectively. Note that we use a different color scale for the 10.7, 18.7, and 37 GHz images to
similar spatial features with high radiances, which are apparent signatures of rain. A rain band
highlight similar spatial features with high radiances, which are apparent signatures of rain. A rain
extending from the center to the northeast can be clearly found in these panels. The 18.7 and 37 GHz
band extending from the center to the northeast can be clearly found in these panels. The 18.7 and
channels are nearly saturated close to the eye of the hurricane because of the influence of
37 GHz channels are nearly saturated close to the eye of the hurricane because of the influence of
precipitation. However, the maximum value of the 10.7 GHz channel near the eye of the hurricane is
precipitation. However, the maximum value of the 10.7 GHz channel near the eye of the hurricane is
approximately 210 K, which is fairly far from saturation. Similar characteristics can be found at lower
approximately 210 K, which is fairly far from saturation. Similar characteristics can be found at lower
frequency, such as 6.8 GHz. In addition, similar results can also be found in the vertically polarized
frequency, such as 6.8 GHz. In addition, similar results can also be found in the vertically polarized
brightness temperatures in the bottom panels of Figure 3. Therefore, we can use the lower
brightness temperatures in the bottom panels of Figure 3. Therefore, we can use the lower frequencies,
frequencies, such as the 6.8 and 10.7 GHz channels, to develop an algorithm for retrieval of high
such as the 6.8 and 10.7 GHz channels, to develop an algorithm for retrieval of high winds under
winds under severe weather conditions.
severe weather conditions.
Remote Sens. 2016, 8, 721
Remote Sens. 2016, 8, 721
8 of 20
8 of 20
a
b
c
d
e
f
Figure
temperatures
measured
by WindSat
orbit orbit
#3510#3510
on 10 on
September
2003, over
Figure3.3.Brightness
Brightness
temperatures
measured
by WindSat
10 September
2003,
hurricane
Isabel
(a)
at
10.7
GHz,
horizontal
polarization;
(b)
at
18.7
GHz,
horizontal
polarization;
(c)
over hurricane Isabel (a) at 10.7 GHz, horizontal polarization; (b) at 18.7 GHz, horizontal polarization;
at(c)37.0
GHz,
horizontal
polarization;
(d) (d)
at 10.7
GHz,
vertical
polarization;
(e)(e)
atat
18.7
at 37.0
GHz,
horizontal
polarization;
at 10.7
GHz,
vertical
polarization;
18.7GHz,
GHz,vertical
vertical
polarization
GHz, vertical
verticalpolarization.
polarization.The
Thered
redcurve
curve
denotes
best
track,
which
polarizationand
and(f)
(f) at
at 37.0 GHz,
denotes
thethe
best
track,
which
was
was
obtained
from
NHC.
obtained
from
the the
NHC.
3.3.
3.3.Wind
WindSpeed
SpeedRetrieval
RetrievalModel
Model
Previous
Previousstudies
studieshave
haveshown
shownthat
thatthe
thesensitivity
sensitivitytotothe
theocean
oceanwind
windatathorizontal
horizontalpolarization
polarizationisis
larger
largerthan
thanthat
thatatatvertical
verticalpolarization
polarization[7,40].
[7,40].The
Thestudy
studyofof[15]
[15]only
onlyused
used6.9
6.9and
and10.7
10.7GHz
GHzhorizontal
horizontal
polarization
(hereafter,
6H
and
10H)
data
from
AMSR-E
to
retrieve
wind
speed
inside
hurricanes.
polarization (hereafter, 6H and 10H) data from AMSR-E to retrieve wind speed inside hurricanes.
However,
the6.9
6.9and
and10.7
10.7
GHz
vertical
polarization
(hereafter,
6 V10V)
anddata
10 V)
in
However, we
we retain
retain the
GHz
vertical
polarization
(hereafter,
6V and
in data
addition
addition
to the horizontal
polarization
data
in this
We the
findvertical
that the
vertical polarization
data
to the horizontal
polarization
data in this
study.
Westudy.
find that
polarization
data also shows
also
shows
the
same
characteristics
as
the
horizontal
polarization
data
described
in
[15].
In
this
study,
the same characteristics as the horizontal polarization data described in [15]. In this study, we provide
we
an algorithm
improved for
algorithm
for retrieving
wind
speed
TCs6V(H)
by using
V(H) and
10V(H)
anprovide
improved
retrieving
wind speed
inside
TCsinside
by using
and6 10V(H)
brightness
brightness
temperatures.
Based
on[15],
the we
study
we
assess the emissivity
wind-induced
emissivity
temperatures.
Based on the
study
also[15],
assess
thealso
wind-induced
changes
at both
changes
at
both
6
V(H)
and
10
V(H)
because
we
retrieve
wind
speed
by
combining
data
from
6V(H) and 10V(H) because we retrieve wind speed by combining data from these channels. these
channels.
Firstly, the brightness temperatures are simulated using Equations (4) and (5). Generally,
Firstly,
the brightness
are simulated
using
Equations
and (5). Generally,
three
three processes
should be temperatures
taken into account:
absorption,
emission,
and(4)
scattering.
The atmospheric
processes
should
be
taken
into
account:
absorption,
emission,
and
scattering.
The
atmospheric
absorption includes three components: water vapor, oxygen, and liquid water in the form of clouds or
absorption
includes
components:
water
vapor, oxygen,
liquid water
the form
of clouds
rain [7]. The
above three
components
also emit
microwave
energyand
according
to thein
ambient
temperature.
orMeanwhile,
rain [7]. Thescattering
above components
also
emit
microwave
energy
according
to
the
ambient
temperature.
must be taken into account unless the microwave wavelength is much larger
Meanwhile,
must bePrevious
taken into
account
unless
the
microwave
wavelength scattering
is much larger
than the sizescattering
of the raindrops.
studies
have
shown
that
the raindrop-induced
can be
than
the size
the raindrops.
Previous
have
shown
raindrop-induced
neglected
foroffrequencies
below
10 GHzstudies
when the
rain
rate isthat
lessthe
than
12 mm/h [15,41].scattering
Scatteringcan
has
be
neglected
for
frequencies
below
10
GHz
when
the
rain
rate
is
less
than
12
mm/h
[15,41].
Scattering
been neglected in our simulation because the selected profiles are in almost clear weather conditions.
has been
in our model
simulation
because
the toselected
profiles
are in almost
clear weather
The neglected
60-level ECMWF
outputs
are used
simulate
the brightness
temperatures
[42] and
conditions.
were generated by the ERA-40 assimilation system, which relies on the 3-dimensional variational
The 60-level
ECMWF
outputs are
used to
simulate
the brightness
temperatures
[42]ocean
and
scheme
described
in [43].model
The sampled
database
gathers
profiles
corresponding
to various
were
generated
ERA-40
assimilation
which
reliesspeed
on theranges
3-dimensional
variational
conditions,
and by
the the
highest
surface
pressure issystem,
1049 hPa.
The wind
from 0.15 to
21.43 m/s,
scheme
described
in
[43].
The
sampled
database
gathers
profiles
corresponding
to
various
and the sea surface temperature ranges from 278.34 to 305.02 K. In addition, the water vapor ocean
ranges
conditions,
and the
surface
is 1049
hPa.
The
wind
ranges from transmittance
0.15 to 21.43
from 4 to 55.64
mm,highest
and the
liquidpressure
water ranges
from
0 to
0.19
mm.speed
The atmospheric
m/s,
and the channels
sea surface
temperature
ranges
from 278.34
to 305.02
K. Inequations
addition, [39,44].
the water
vapor
at different
is computed
from
the standard
radiative
transfer
Likewise,
ranges
from
4
to
55.64
mm,
and
the
liquid
water
ranges
from
0
to
0.19
mm.
The
atmospheric
the downwelling and upwelling atmospheric brightness temperatures are also calculated from the
transmittance
at different
computed
from
standard
transfer
equations
standard radiative
transfer channels
equations is
[39,44].
Note that
thethe
oxygen,
waterradiative
vapor, and
cloud liquid
water
[39,44].
Likewise,
the
downwelling
and
upwelling
atmospheric
brightness
temperatures
are
also
absorption coefficient use O2-MPM93 expression [45], H2O-PWR98 expression [46], and clw-MPM93
calculated from the standard radiative transfer equations [39,44]. Note that the oxygen, water vapor,
and cloud liquid water absorption coefficient use O2-MPM93 expression [45], H2O-PWR98
Remote Sens. 2016, 8, 721
9 of 20
Remote Sens. 2016, 8, 721
9 of 20
expression [45], respectively. In addition, the ocean surface emissivity and the reflectivity are calculated
expression [46], and clw-MPM93 expression [45], respectively. In addition, the ocean surface
with the RTTOV radiative transfer model [47], and the correction factors are derived from RTTOV
emissivity and the reflectivity are calculated with the RTTOV radiative transfer model [47], and the
v11.3. The cosmic background radiation is approximately 2.7 K.
correction factors are derived from RTTOV v11.3. The cosmic background radiation is approximately
To account for wind speed in ocean emission, one parameter is defined as 6(10)P* (P = V, H),
2.7 K.
which represents
between
theemission,
WindSatone
6(10)P
and calm
ocean emission,
To accountthe
fordifference
wind speed
in ocean
parameter
is defined
as 6(10)P* corrected
(P = V, H),for
atmospheric
effects. This
parameter
represents
the quantity
ofand
ocean
microwave
emission
changed
which represents
the difference
between
the WindSat
6(10)P
calm
ocean emission,
corrected
forby
ocean
wind
at
6(10)P,
which
is
defined
by
Equation
(8).
atmospheric effects. This parameter represents the quantity of ocean microwave emission changed
by ocean wind at 6(10)P, which is defined by Equation (8).
6 (10) P∗ = WindSat_6 (10) P − Atmos_6 (10) P − CalmOcean_6 (10) P
6 (10 ) P* = WindSat _ 6 (10 ) P − Atmos _ 6 (10 ) P − CalmOcean _ 6 (10 ) P
CalmOcean_6 (10) P = SST × 1 − rp
CalmOcean _ 6 (10 ) P = SST × (1 − rp )
(8)
(8)
(9)
(9)
where the WindSat_6(10)P is the brightness temperature measurement for the 6(10) GHz channel at
P-polarization
(P = V, H) and isCalmOcean_6(10)P
is the ocean
emission for
forthe
6(10)
GHz
channel
under
where the WindSat_6(10)P
the brightness temperature
measurement
6(10)
GHz
channel
at
calmP-polarization
ocean conditions.
The
SST
used
here
is
from
ECMWF.
In
addition,
Atmos_6(10)P
is
calculated
(P = V, H) and CalmOcean_6(10)P is the ocean emission for 6(10) GHz channel underby
referring
to [48].
calm ocean
conditions. The SST used here is from ECMWF. In addition, Atmos_6(10)P is calculated
by
referring
[48].
Figure
4a,btoillustrate
the relationships between 6(10)V* and ECMWF wind speed. The 6(10)V*
Figure
4a,b
illustrate
the
relationships
ECMWF
windofspeed.
6(10)V* at
decreases initially because the
shielding
effectbetween
has been6(10)V*
found and
in the
calculation
oceanThe
emissivity
decreases
initially
because
the
shielding
effect
has
been
found
in
the
calculation
of
ocean
emissivity
V-polarization and then increases with increasing wind speed. The sensitivity of 6(10)V* to wind speed
at V-polarization
and then
increases
withthe
increasing
windisspeed.
of 6(10)V*
windthe
is approximately
0.37(0.5)
K/(m/s)
when
wind speed
from The
8 to sensitivity
20 m/s. Figure
4d,eto
show
speed
is
approximately
0.37(0.5)
K/(m/s)
when
the
wind
speed
is
from
8
to
20
m/s.
Figures
4d
and
4e
relationships between 6(10)H* and ECMWF wind speed. The 6(10)H* increases with increasing
show
the
relationships
between
6(10)H*
and
ECMWF
wind
speed.
The
6(10)H*
increases
with
wind speed, which is different from 6(10)V*. The sensitivity of 6(10)H* to wind speed is better than
increasing wind speed, which is different from 6(10)V*. The sensitivity of 6(10)H* to wind speed is
that of 6(10)V*, and it is approximately 0.85(0.93) K/(m/s). The previous study [15] showed that the
better than that of 6(10)V*, and it is approximately 0.85(0.93) K/(m/s). The previous study [15] showed
sensitivity of 6H* to wind speed is approximately 1 K/(m/s), which is higher than our calculated result.
that the sensitivity of 6H* to wind speed is approximately 1 K/(m/s), which is higher than our
The diverse data resources may result in this difference. On the contrary, our calculated result of 6H* is
calculated result. The diverse data resources may result in this difference. On the contrary, our
higher than the experimental result [49], which indicates that the sensitivity at approximately 6–7H is
calculated result of 6H* is higher than the experimental result [49], which indicates that the sensitivity
approximately
0.5 K/(m/s),
though the
arethough
limitedthe
to experiments
ground-based
at approximately
6–7H is even
approximately
0.5experiments
K/(m/s), even
areobservations.
limited to
Figure
4c
illustrate
the
relationships
between
the
6V*
and
10V*.
When
the
wind
speed
is from
0 to
ground-based observations. Figure 4c illustrate the relationships between the 6V* and 10V*.
When
20 m/s,
the ratio
to010V*
a constant
is 10V*
approximately
0.82.
Theisratio
of 6H* and
10H*
the wind
speedofis6V*
from
to 20 is
m/s,
the ratio which
of 6V* to
is a constant
which
approximately
0.82.
has The
similar
characteristics
(Figure
4f),
and
the
value
is
approximately
0.92.
These
results
are
in
good
ratio of 6H* and 10H* has similar characteristics (Figure 4f), and the value is approximately 0.92.
agreement
with the
previous
results [15,28].
These results
are in
good agreement
with the previous results [15,28].
Figure
4. (a)
Relationshipbetween
between 6V*
wind
speed;
(b) Relationship
between
10V* and
Figure
4. (a)
Relationship
6V*and
andECMWF
ECMWF
wind
speed;
(b) Relationship
between
10V*
speed;
(c) (c)
Relationship
between
6V* 6V*
and and
10V*;10V*;
(d) Relationship
between
6H* and
andECMWF
ECMWFwind
wind
speed;
Relationship
between
(d) Relationship
between
6H*
ECMWF
wind
speed;
(e)
Relationship
between
10H*
and
ECMWF
wind
speed;
(f)
Relationship
and ECMWF wind speed; (e) Relationship between 10H* and ECMWF wind speed; (f) Relationship
between
10H*.
between
6H*6H*
andand
10H*.
Remote Sens. 2016, 8, 721
10 of 20
Making use of the fact that the ratio between 6H* and 10H* is a constant, Shibata [15] established
a wind speed retrieval algorithm by assuming that this ratio still holds at higher wind speeds and under
rainy conditions. Based on the Shibata method and the above discussions, we develop an improved
algorithm with two new parameters known as W6V and W6H. The definitions of W6V and W6H are
similar to the previous study [15]:
W6H
W6V
= EF∗ /fac1 = (6H− − 6HE− )/fac1
= (6H− − c1 × 10H− + a1 × c1 − b1 ) × sl1 /(sl1 − c1 )/fac1
(10)
sl1 = d1 + e1 × (10HE− − a1 )
(11)
fac1 = 1 − f1 × (10HE− − a1 )
(12)
6(10)H− = WindSat_6(10)H − CalmOcean_6(10)H
(13)
= EF∗ /fac2 = (6V− − 6VE− )/fac2
= (6V− − c2 × 10V− + a2 × c2 − b2 ) × sl2 /(sl2 − c2 )/fac2
(14)
sl2 = d2 + e2 × (10VE− − a2 )
(15)
fac2 = 1 − f2 × (10VE− − a2 )
(16)
6(10)V− = WindSat_6(10)V − CalmOcean_6(10)V
(17)
where the WindSat_6(10)P and CalmOcean_6(10)P are the same as Equations (8) and (9). The definition
of other parameters in Equations (10) to (13) can be found in Figure 5. The brightness temperatures of
group A (Figure 5a) are the results under calm ocean conditions. The brightness temperatures of group
B (Figure 5a) are the results under rough ocean conditions. In Figure 5a, the relationship between 6H
and 10H is slightly nonlinear, but we have approximated it as linear for simplicity. By calculating
6(10) H¯, we can obtain Figure 5b, in which line A is approximated as a linear line passing through
the point O(a1 ,b1 ) with the slope c1 . Point F in Figure 5b corresponds to an arbitrary point on line B
in Figure 5a. Then, we can obtain an intersecting point E made by two lines OP and EF in Figure 5b.
The slope of the line EF is sl1 . Finally, we can obtain the parameter “W6H” (unit: Kelvin) based on the
length of EF* divided by the atmospheric effect (=fac1 ). A detailed description can be found in [15].
The parameters for V-polarization in Equations (14) to (17) are similar to those of H-polarization in
Equations (10) to (13).
Finally, we propose a new algorithm that relates W6V and W6H to wind speed (m/s):
WS = 0.2034 × W6H + 0.01 × W6V + 15.3
if W6H < 20
WS = 0.3017 × (W6H − 20) + 0.2 × (W6V − 30) + 22.65 if 20 ≤ W6H < 30
WS = 0.3 × (W6H − 30) + 0.4 × (W6V − 40) + 32.54
if W6H ≥ 30
The coefficients a1 ~f1 and a2 ~f2 in Equations (10) to (17) are listed in Table 2.
(18)
Remote Sens. 2016, 8, 721
Remote Sens. 2016, 8, 721
11 of 20
11 of 20
H
H
E
E
−
(a)
−
(b)
Figure 5. (a) Brightness temperatures of 6H and 10H: group A for calm ocean conditions and group
Figure
5. (a) Brightness temperatures of 6H and 10H: group A for calm ocean conditions and group B
B for rough ocean conditions; (b) Thematic figure drawn from (a). The two figures can be found in
for rough ocean conditions; (b) Thematic figure drawn from (a). The two figures can be found in [15].
[15].
Table 2. Model coefficients in Equations (10) to (17).
Table 2. Model coefficients in Equations (10) to (17).
a1
b1 b1
a1
14.171814.1718 6.0173
6.0173
a2
b2
a
b
17.0839 2
3.1643 2
17.0839
3.1643
c1 c1
0.3284
0.3284
c2
c
0.43302
0.4330
dd11
0.9521
0.9521
d2
d2
0.9529
0.9529
e1 e1
0.0100
0.0100
e2
e2
0.0011
0.0011
f1
0.00100.0010
f2
f2
0.0018
0.0018
f1
4. Results
4. Results
4.1. Brightness Temperature 6(10)P¯ Comparisons
4.1. Brightness Temperature 6(10)P¯ Comparisons
Before calculating the fitted coefficients of the model, we must estimate the 6(10) P¯ (P = V, H)
with
Equation
(13) and
(17). We of
first
to calculate
oceanthe
TBs
under
the
calm
Before
calculating
theEquation
fitted coefficients
theneed
model,
we must the
estimate
6(10)
P¯ (P
= V,
H) sea
with
conditions,
which
are
a
function
of
frequency,
polarization,
incidence
angle,
SST
and
salinity.
The
Equation (13) and Equation (17). We first need to calculate the ocean TBs under the calm sea conditions,
SST derived
from theofReynolds
weekly
analysis was
used toangle,
estimate
ocean
microwave
emission
which
are a function
frequency,
polarization,
incidence
SSTthe
and
salinity.
The SST
derived
at 6(10)
underweekly
the calm
sea conditions
with
the Kleinthe
andocean
Swift microwave
model [50]. For
the salinity
effect,
from
the GHz
Reynolds
analysis
was used
to estimate
emission
at 6(10)
GHz
the
previous
study
[15]
demonstrated
that
the
sensitivity
of
TBs
at
6.8
GHz
H-polarization
is
0.003
under the calm sea conditions with the Klein and Swift model [50]. For the salinity effect, the previous
◦C
K/PSU
at demonstrated
0 °C SST and −0.039
K/PSU
at 30 °CofSST
salinity
range of 30 to 35
values
study
[15]
that the
sensitivity
TBsfor
at a6.8
GHz H-polarization
is PSU.
0.003The
K/PSU
at 0for
◦ C SST
10.7and
GHz
H-polarization
K/PSU
−0.012range
K/PSU,
salinity
effectGHz
is
SST
−0.039
K/PSU atare
300.006
for aand
salinity
of respectively.
30 to 35 PSU.Thus,
The the
values
for 10.7
also
negligible
in
our
study.
H-polarization are 0.006 K/PSU and −0.012 K/PSU, respectively. Thus, the salinity effect is also
Figure
6 shows
negligible
in our
study.the brightness temperatures of 6H¯ (V¯) and 10H¯ (V¯) inside and around
hurricanes.
casesthe
in which
the center
of the hurricane
contains
land10H¯
areas(V¯)
wereinside
removed
Figure 6The
shows
brightness
temperatures
of 6H¯
(V¯) and
andbecause
around
6H¯
(V¯)
and
10H¯
(V¯)
cannot
be
evaluated
over
land.
Figure
6a
shows
the
WindSat
6H¯
andbecause
10H¯
hurricanes. The cases in which the center of the hurricane contains land areas were removed
from hurricane Fabian on 3 September 2003, and the maximum wind speed is 47 m/s. Figure 6b
6H¯ (V¯) and 10H¯ (V¯) cannot be evaluated over land. Figure 6a shows the WindSat 6H¯ and 10H¯
displays the WindSat 6H¯ and 10H¯ on 31 August 2004, with a maximum wind speed of 51 m/s for
from hurricane Fabian on 3 September 2003, and the maximum wind speed is 47 m/s. Figure 6b
hurricane Frances. Figure 6c presents the WindSat 6H¯ and 10H¯ on 9 July 2005, with a maximum
displays the WindSat 6H¯ and 10H¯ on 31 August 2004, with a maximum wind speed of 51 m/s for
wind speed of 51 m/s for hurricane Dennis; Figure 6d–f illustrate the results of the vertically polarized
hurricane Frances. Figure 6c presents the WindSat 6H¯ and 10H¯ on 9 July 2005, with a maximum
channels. In Figure 6a, the data marked by arrow A correspond to the outer area around Fabian, and
wind speed of 51 m/s for hurricane Dennis; Figure 6d–f illustrate the results of the vertically polarized
the data marked by B correspond to the area inside Fabian. The situation is similar for the other cases.
channels. In Figure 6a, the data marked by arrow A correspond to the outer area around Fabian, and
In addition, the arrow B moves upward as the wind speed increases, and the above results are in
the data marked by B correspond to the area inside Fabian. The situation is similar for the other cases.
good agreement with the previous study [15].
In addition, the arrow B moves upward as the wind speed increases, and the above results are in good
agreement with the previous study [15].
Remote Sens. 2016, 8, 721
Remote
RemoteSens.
Sens.2016,
2016,8,
8,721
721
12 of 20
12
12of
of20
20
60
60
60
60
-
-
6H - (K)
(K)
6H
70
70
80
80
6H- (K)
(K)
6H
100
100
40
40
50
100
50
100
-10H
10H (K)
(K)
20
20
20
20
150
150
40
40
40
40
30
30
6V - (K)
(K)
6V
60
60
-
-
6V- (K)
(K)
6V
40
40
30
30
20
20
00
50
50
20
20
40
40
60
60 80
80 100
100 120
120
-10H
10H (K)
(K)
20
20
10
10
00
00
50
50
-10V
10V (K)
(K)
100
100
00
20
20
40
60
40
60
-10V
10V (K)
(K)
80
80
¯ and
¯ ¯and
Figure
WindSat
6H
10H
for
three
hurricanes:
Figure
6. Relationships
Relationships
of
WindSat6H
6H
and10H
10H¯¯¯for
forthree
threehurricanes:
hurricanes: (a)
(a)
Fabian
(2003);
(b)
Frances
Figure
6. 6.
Relationships
ofof
WindSat
(a)Fabian
Fabian(2003);
(2003);(b)
(b)Frances
Frances
¯¯ and 10V¯¯¯ for three hurricanes: (d) Fabian
¯
(2004);
and
(c)
Dennis
(2005).
Relationships
of
WindSat
6V
(2004);
Dennis
(2005).
RelationshipsofofWindSat
WindSat6V
6V and
and 10V
10V for
for three
three hurricanes:
(2004);
andand
(c) (c)
Dennis
(2005).
Relationships
hurricanes:(d)
(d)Fabian
Fabian
(2003);
Frances
(2004);
and
Dennis
(2005).
(2003);
(e)
Frances
(2004);
and
(f)
Dennis(2005).
(2005).
(2003);
(e)(e)
Frances
(2004);
and
(f)(f)
Dennis
4.2.
4.2. Wind
Wind Speed
Speed Retrieval
Retrieval and
and Validation
Validation
4.2. Wind Speed Retrieval and Validation
The
The new
new model
model is
is established
established by
by using
using 4862
4862 matchups
matchups between
between WindSat
WindSat TBs
TBs and
and H*wind
H*wind
The new model is established by using 4862 matchups between WindSat TBs and H*wind analysis
analysis
data
for
Fabian
(2003),
Frances
(2004),
and
Dennis
(2005).
We
have
checked
that
other
analysis data for Fabian (2003), Frances (2004), and Dennis (2005). We have checked that othergroups
groups
data for Fabian (2003), Frances (2004), and Dennis (2005). We have checked that other groups of
of
of hurricanes
hurricanes give
give similar
similar fitted
fitted coefficients
coefficients for
for wind
wind speed
speed retrieval.
retrieval. The
The results
results show
show that
that the
the mean
mean
hurricanes give similar fitted coefficients for wind speed retrieval. The results show that the mean bias
bias
bias and
and RMS
RMS difference
difference of
of the
the new
new model
model (Figure
(Figure 7a)
7a) are
are 0.07
0.07 m/s
m/s and
and 2.63
2.63 m/s,
m/s, respectively.
respectively.
and
RMS difference also
of the new model
(Figure
7a) are the
0.07 m/s and
2.63 m/s,
respectively.with
Furthermore,
Furthermore,
Furthermore, we
we also compare
compare the
the new
new model
model to
to the model
model of
of Shibata
Shibata (2006)
(2006) model
model with the
the old
old
wecoefficients
also
compare
the
new
model
to
the
model
of
Shibata
(2006)
model
with
the
old
coefficients
[15]
[15]
and
the
new
coefficients
derived
from
our
matchups,
respectively.
The
performance
coefficients [15] and the new coefficients derived from our matchups, respectively. The performance
and
derived from
our
matchups,
respectively.
Theof
of the new
model
of
the
new
model
The
bias
the
with
the
ofthe
thenew
newcoefficients
model is
is improved.
improved.
The mean
mean
bias and
and RMS
RMS difference
difference
ofperformance
the Shibata
Shibata method
method
with
the
is improved.
The
mean
bias
and
RMS
difference
of
the
Shibata
method
with
the
new
coefficients
new
new coefficients
coefficients (Figure
(Figure 7b)
7b) are
are −0.22
−0.22 m/s
m/s and
and 2.81
2.81 m/s,
m/s, respectively,
respectively, which
which are
are better
better than
than those
those
(Figure
7b) are
−the
0.22
m/s and
2.81 using
m/s,
whichthat
are better
than those
retrieved
with the
retrieved
with
model
the
were
from
the
retrieved
with
the Shibata
Shibata
model
using respectively,
the old
old coefficients
coefficients
that
were derived
derived
from
the simulations
simulations
Shibata
model
using
the
old
coefficients
that
were
derived
from
the
simulations
(with
mean
bias
and
(with
mean
bias
and
RMS
difference
of
1.23
m/s
and
5.13
m/s,
respectively).
In
addition,
the
error
(with mean bias and RMS difference of 1.23 m/s and 5.13 m/s, respectively). In addition, the
error
RMS
difference
of
1.23
m/s
and
5.13
m/s,
respectively).
In
addition,
the
error
increases
with
increasing
increases
increases with
with increasing
increasing wind
wind speed
speed for
for both
both the
the new
new model
model and
and the
the Shibata
Shibata model.
model. Through
Through the
the
wind
speed
for bothwe
thecan
new
model and
the Shibata
model. Through
above discussion,
wethe
can
above
discussion,
conclude
the
V-polarization
channels
contribute
to
above
discussion,
we
can
conclude
the added
added
V-polarization
channelsthe
contribute
to improving
improving
the
conclude
the
added V-polarization channels contribute to improving the retrieval performance.
retrieval
performance.
retrieval
performance.
(a)
(a)
(b)
(b)
Figure
wind
speed
three
hurricanes
using
our
new
Figure
7. Retrieved
Retrieved
wind
speedofof
ofthree
threehurricanes
hurricanesusing
usingour
our new
new method
method and
and
the
Shibata
method
Figure
7. 7.
Retrieved
wind
speed
method
andthe
theShibata
Shibatamethod
method
with
the
new
coefficients
versus
H*wind
analysis
data.
(a)
Our
new
method
and
(b)
the
method
of
with
the
new
coefficients
versus
H*wind
analysis
data.
(a)
Our
new
method
and
(b)
the
method
with the new coefficients versus H*wind analysis data. (a) Our new method and (b) the methodofof
Shibata
(2006)
with
the
new
coefficients.
Shibata
(2006)
with
new
coefficients.
Shibata
(2006)
with
thethe
new
coefficients.
4.2.1.
4.2.1. Comparison
Comparison with
with H*wind
H*wind Analysis
Analysis Data
Data
4.2.1. Comparison with H*wind Analysis Data
Figure
Figure 88 presents
presents the
the retrieved
retrieved wind
wind speed
speed using
using the
the new
new model
model versus
versus the
the corresponding
corresponding
Figure 8 presents
the retrieved
wind speed using
the new model
versus the
corresponding
H*wind
H*wind
H*wind analysis
analysis data
data for
for each
each selected
selected hurricane.
hurricane. The
The mean
mean bias
bias and
and RMS
RMS error
error are
are given
given in
in the
the top
top
analysis
data
for
each
selected
hurricane.
The
mean
bias
and
RMS
error
are
given
in
the
top
of
each
of
each
single
panel.
For
wind
speeds
above
20
m/s,
the
mean
bias
and
overall
RMS
difference
of
the
of each single panel. For wind speeds above 20 m/s, the mean bias and overall RMS difference of the
single
For
wind
speeds
above
m/s, the mean
and overall
RMS
of the
new
new
model
m/s
and
m/s,
respectively,
for
hurricanes
from
to
2010.
biases
newpanel.
modelare
are0.04
0.04
m/s
and2.75
2.75
m/s,20
respectively,
for15
15bias
hurricanes
from2003
2003
todifference
2010.Positive
Positive
biases
model
are
0.04
m/s
and
2.75
m/s,
respectively,
for
15
hurricanes
from
2003
to
2010.
Positive
biases
are
are found
found in
in some
some hurricanes,
hurricanes, including
including Dennis,
Dennis, Katrina,
Katrina, Rita,
Rita, Wilma,
Wilma, Helene,
Helene, Ike,
Ike, Bill,
Bill, and
and Ida.
Ida. For
For
are found in some hurricanes, including Dennis, Katrina, Rita, Wilma, Helene, Ike, Bill, and Ida.
Remote Sens. 2016, 8, 721
13 of 20
Remote Sens. 2016, 8, 721
13 of 20
For hurricane Bill, in particular, the new model largely overestimates wind speeds, with a mean bias of
hurricane Bill, in particular, the new model largely overestimates wind speeds, with a mean bias of
2.89 m/s. On the contrary, the largest negative bias (−1.45 m/s) is found in hurricane Bertha. On the
2.89 m/s. On the contrary, the largest negative bias (−1.45 m/s) is found in hurricane Bertha. On the
whole,
the retrieved results have a good agreement with the H*wind analysis wind speed.
whole, the retrieved results have a good agreement with the H*wind analysis wind speed.
It isItchallenging
to to
retrieve
wind
conditions,especially
especially
typhoons
is challenging
retrieve
windspeeds
speedsunder
undersevere
severe weather
weather conditions,
inin
typhoons
andand
hurricanes.
Rain
not
only
increases
the
atmospheric
attenuation
but
also
changes
the
sea
surface
hurricanes. Rain not only increases the atmospheric attenuation but also changes the sea surface
roughness
in aincomplicated
manner.
modelbrightness
brightnesstemperatures
temperatures
roughness
a complicated
manner.Thus,
Thus,ititisisdifficult
difficult to
to accurately
accurately model
under
rainy
conditions.
Hence,
the
statistical
algorithm
is
a
relatively
good
candidate
that
can
used
under rainy conditions. Hence, the statistical algorithm is a relatively good candidate that can
bebe
used
to retrieve
wind
speed
under
severe
general,the
thenew
newwind
windspeed
speedretrieval
retrieval
to retrieve
wind
speed
under
severeweather
weatherconditions.
conditions. In general,
algorithm
under
rain
gives
encouragingaccuracy
accuracycompared
compared to the H*wind
algorithm
under
rain
gives
anan
encouraging
H*windwind
windspeed.
speed.
Hwind wind speed (m/s)
Bias =-0.68, RMS =2.73
50
(a)
40
30
20
20
30
40
50
Retrieved wind speed (m/s)
Bias =1.28, RMS =2.60
45
Bias =0.45, RMS =1.51
(e)
50
Hwind Ws (m/s)
Hwind wind speed (m/s)
50
40
35
30
25
20
20
30
40
Retrieved wind speed (m/s)
20
Hwind wind speed (m/s)
Hwind wind speed (m/s)
40
30
20
Hwind wind speed (m/s)
Hwind wind speed (m/s)
40
30
20
Bias =2.89, RMS =3.67
Hwind wind speed (m/s)
Hwind wind speed (m/s)
45
40
35
30
25
20
15
20
30
40
Retrieved wind speed (m/s)
1.5
1
0
50
2
40
35
30
25
20
15
20
30
40
Retrieved wind speed (m/s)
50
40
35
30
25
20
20
30
40
Retrieved wind speed (m/s)
45
50
Bias =0.55, RMS =2.20
50
(m)
(i)
(k)
15
20
30
40
50
Retrieved wind speed (m/s)
50
45
50
Bias =-1.45, RMS =2.61
50
(j)
45
30
40
Retrieved Ws (m/s)
Bias =0.71, RMS =2.14
50
(h)
20
30
40
50
Retrieved wind speed (m/s)
50
30
50
Bias =1.82, RMS =2.26
Bias =-0.35, RMS =3.28
40
20
15
50
(f)
(n)
40
35
30
25
20
15
20
30
40
Retrieved wind speed (m/s)
4
6
50
8
10
12
Figure
8. The
comparisons
betweenH*wind
H*windwind
windspeed
speed and
and the
our
new
Figure
8. The
comparisons
between
the retrieved
retrievedwind
windspeed
speedwith
with
our
new
model
using
WindSat
6 and
GHz
brightnesstemperatures
temperatures for
for each
(a)(a)
Fabian;
model
using
WindSat
6 and
1010
GHz
brightness
each selected
selectedhurricane:
hurricane:
Fabian;
(b) Isabel;
(c) Frances;
(d) Ivan;
(e) Dennis;
(f) Katrina;
(g) Rita;
(h) Wilma;
(i) Helene;
(j) Felix;
(k)
(b) Isabel;
(c) Frances;
(d) Ivan;
(e) Dennis;
(f) Katrina;
(g) Rita;
(h) Wilma;
(i) Helene;
(j) Felix;
(k) Bertha;
Bertha;
(l)
Ike;
(m)
Bill;
(n)
Ida;
and
(o)
Igor.
(l) Ike; (m) Bill; (n) Ida; and (o) Igor.
Remote Sens. 2016, 8, 721
14 of 20
4.2.2. Comparison with Remote Sensing Systems (RSS) All-Weather Data
A global all-weather wind speed product has been developed using the WindSat TBs [10], which is
a good candidate for comparison to our retrieved wind speed results based on WindSat TBs inside
hurricanes. Table 3 illustrates the retrieved wind speed using our new algorithm and the RSS global
algorithm, respectively. The overall RMS difference of wind speed between our new model and H*wind
analysis data is lower than that between the RSS wind speed and H*wind analysis data. In addition,
results are similar for each individual case expect for hurricane Bill. All biases are negative between
the RSS wind speed and H*wind analysis data. That means the RSS wind speed underestimates wind
speed on the whole, especially for high wind speeds. The mean bias is −2.53 m/s between them.
Compared to the H*wind analysis data, the retrieved wind speed using the new method overestimates
wind speed slightly, and the mean bias is 0.04 m/s for the selected 15 hurricanes.
Table 3. Statistics of errors between the new retrieved wind speed and H*wind wind speed and
between RSS all-weather wind speed and H*wind wind speed for H*wind wind speeds above 20 m/s
for the selected hurricanes from 2003 to 2010.
Hurricane Name
Ours
RSS
Bias (m/s)
RMS (m/s)
Bias (m/s)
RMS (m/s)
Fabian
Isabel
Frances
Ivan
Dennis
Katrina
Rita
Wilma
Helene
Felix
Bertha
Ike
Bill
Ida
Igor
−0.68
−1.04
−0.30
−0.47
1.28
0.45
0.65
1.82
0.71
−0.35
−1.45
0.87
2.89
0.55
−0.66
2.73
3.34
2.92
2.71
2.60
1.51
1.97
2.26
2.14
3.28
2.61
3.20
3.67
2.20
2.29
−3.02
−3.25
−3.03
−2.68
−2.46
−2.32
−2.38
−1.65
−1.86
−2.80
−4.30
−2.30
−1.40
−1.50
−1.73
4.26
4.39
4.66
3.65
3.89
3.68
3.20
2.90
3.25
4.21
5.40
4.04
3.25
4.00
3.29
All together
0.04
2.75
−2.53
3.91
4.2.3. Comparison with SFMR Measurements
The SFMR measurements are compared with our retrieved wind speed results, although these
data can only provide point observations along the aircraft flight track. The 30-min window is
utilized to collocate the matchups between the WindSat data and SFMR measurements. Since the
SFMR measurements are at a high spatial resolution of 1.5 km, a method similar to that described
in section 2.5 is also performed. Finally, we obtained 938 wind speed matchups for the selected
15 hurricanes. Figure 9 suggests that the derived wind speed has a relatively good agreement with the
SFMR measurements, with a mean bias of −0.68 m/s and an RMS difference of 3.14 m/s. The general
trend is that the retrieved wind speed errors increase with increasing SFMR rain rate.
Remote Sens. 2016, 8, 721
Remote Sens. 2016, 8, 721
50
8
40
6
Number = 938
45 Bias = -0.68 m/s
RMS = 3.14 m/s
Error (m/s)
SFMR wind speed (m/s)
15 of 20
15 of 20
35
30
25
4
2
20
15
20
30
40
Retrieved wind speed (m/s)
(a)
50
0
0
5
10
15
SFMR rain rate (mm/h)
(b)
Figure 9. (a) The comparison between SFMR wind speeds and the retrieved wind speeds with our
Figure 9. (a) The comparison between SFMR wind speeds and the retrieved wind speeds with our new
new method; (b) Errors of the retrieved wind speeds versus SFMR rain rate.
method; (b) Errors of the retrieved wind speeds versus SFMR rain rate.
4.3. Case Studies
4.3. Case Studies
As a case study (denoted as Rita-1), two different models are used to retrieve wind speed using
As a case study (denoted as Rita-1), two different models are used to retrieve wind speed
the WindSat TBs for hurricane Rita on 22 September 2005, at 1330 UTC. Figure 10a presents the
using the WindSat TBs for hurricane Rita on 22 September 2005, at 1330 UTC. Figure 10a presents
H*wind analysis wind speed, and Figure 10b shows the retrieved wind speed using our new model.
the H*wind analysis wind speed, and Figure 10b shows the retrieved wind speed using our new
Figure 10c shows the retrieved wind speed using the RSS global algorithm, and Figure 10d illustrates
model. Figure 10c shows the retrieved wind speed using the RSS global algorithm, and Figure 10d
the corresponding RSS rain rate. Figure 10e shows the difference between the retrieved wind speed
illustrates the corresponding RSS rain rate. Figure 10e shows the difference between the retrieved
and the H*wind analysis data. The difference between the RSS wind speed and the H*wind analysis
wind speed and the H*wind analysis data. The difference between the RSS wind speed and the
data is presented in Figure 10f. In addition, the corresponding W6H and W6V are presented in Figure
H*wind analysis data is presented in Figure 10f. In addition, the corresponding W6H and W6V are
10g,h, respectively. In this case, the H*wind analysis wind speed is obtained for 22 September 2005,
presented in Figure 10g,h, respectively. In this case, the H*wind analysis wind speed is obtained for
at 1330 UTC. At that time, the center of typhoon’s eye was located at 25.197°N and −88.531°E.
22 September 2005, at 1330 UTC. At that time, the center of typhoon’s eye was located at 25.197◦ N and
However,
the center of typhoon’s eye was located at 25.141°N and −88.237°E at the time of the
−88.531◦ E. However, the center of typhoon’s eye was located at 25.141◦ N and −88.237◦ E at the time of
WindSat observation, which is obtained from the NHC. Therefore, we shifted the H*wind analysis
the WindSat observation, which is obtained from the NHC. Therefore, we shifted the H*wind analysis
wind speed so that the eye of the H*wind analysis coincides with eye of the WindSat measurement.
wind speed so that the eye of the H*wind analysis coincides with eye of the WindSat measurement.
Comparing W6H with W6V, we find that W6H seems to have a better correlation than W6V. The
Comparing W6H with W6V, we find that W6H seems to have a better correlation than W6V.
distribution of W6H is similar to that of wind speed. The performance of the new model is satisfied,
The distribution of W6H is similar to that of wind speed. The performance of the new model is
and the distribution of retrieved wind speed is very close to the H*wind analysis wind speed, with
satisfied, and the distribution of retrieved wind speed is very close to the H*wind analysis wind
the mean bias of 0.79 m/s and RMS difference of 1.79 m/s. As for RSS wind speeds, the influence of
speed, with the mean bias of 0.79 m/s and RMS difference of 1.79 m/s. As for RSS wind speeds,
rain is obvious. On the contrary, our new model can reduce the influence of rain to some extent, and
the influence of rain is obvious. On the contrary, our new model can reduce the influence of rain to
the retrieved results (Figure 10e) are in good agreement with the H*wind analysis wind speed. In
some extent, and the retrieved results (Figure 10e) are in good agreement with the H*wind analysis
general, the RSS global wind speed underestimates the wind speed in most regions, and the retrieved
wind speed. In general, the RSS global wind speed underestimates the wind speed in most regions,
results are vulnerable to rain (Figure 10f). In addition, it is seen that the retrieved wind speed
and the retrieved results are vulnerable to rain (Figure 10f). In addition, it is seen that the retrieved
overestimates wind speed in the center of the eye for our retrieved results and RSS wind speeds. This
wind speed overestimates wind speed in the center of the eye for our retrieved results and RSS wind
may be because there are still strong waves that increase the emission from ocean surface, even
speeds. This may be because there are still strong waves that increase the emission from ocean surface,
though the wind speed is relatively low in the eye of the typhoon.
even though the wind speed is relatively low in the eye of the typhoon.
In a second case study (denoted as Rita-2), the two different models are also utilized to retrieve
In a second case study (denoted as Rita-2), the two different models are also utilized to retrieve
wind speed with the WindSat TBs for hurricane Rita on 22 September 2005, at 0130 UTC. Figure 11b
wind speed with the WindSat TBs for hurricane Rita on 22 September 2005, at 0130 UTC. Figure 11b
illustrates the retrieved wind speed using our new model. Figure 11c shows the retrieved wind speed
illustrates the retrieved wind speed using our new model. Figure 11c shows the retrieved wind speed
using the RSS global algorithm, and Figure 11d shows the corresponding RSS rain rate. Figure 11e–f
using the RSS global algorithm, and Figure 11d shows the corresponding RSS rain rate. Figure 11e–f
present the difference between the retrieved wind speeds and the H*wind analysis data. In addition,
present the difference between the retrieved wind speeds and the H*wind analysis data. In addition,
Figures 11g,h show the corresponding W6H and W6V results, respectively. In this case, the center of
Figure 11g,h show the corresponding W6H and W6V results, respectively. In this case, the center
the typhoon’s eye is located at 24.536°N and −87.238°E for the H*wind analysis data; however, the
of the typhoon’s eye is located at 24.536◦ N and −87.238◦ E for the H*wind analysis data; however,
center of typhoon’s eye is located at 24.535°N and −86.808°E at the time of the WindSat observation.
the center of typhoon’s eye is located at 24.535◦ N and −86.808◦ E at the time of the WindSat observation.
Hence, we shifted the H*wind analysis wind speed so that the eye of the H*wind analysis wind speed
Hence, we shifted the H*wind analysis wind speed so that the eye of the H*wind analysis wind speed
coincides with eye of the WindSat measurement. The results show that the performance of the new
coincides with eye of the WindSat measurement. The results show that the performance of the new
model is satisfied, and the mean bias and RMS difference of the new model are 0.63 m/s and 2.38 m/s,
model is satisfied, and the mean bias and RMS difference of the new model are 0.63 m/s and 2.38 m/s,
respectively. Near the south of the typhoon’s eye, our new model seems to underestimate wind speed
when the rain rate is relatively low, and similar results are also found in the RSS global algorithm.
Remote Sens. 2016, 8, 721
16 of 20
respectively. Near the south of the typhoon’s eye, our new model seems to underestimate wind speed
when
rain
rate
is relatively low, and similar results are also found in the RSS global algorithm.
Remotethe
Sens.
2016,
8, 721
16 of 20
Remote Sens. 2016, 8, 721
16 of 20
(a)
(a)
(m/s)
(m/s)
(m/s)
(m/s)
(b)
(b)
(m/s)
(m/s)
(e)
(e)
(m/s)
(m/s)
(c)
(c)
(m/s)
(m/s)
(f)
(f)
(mm/h)
(mm/h)
(d)
(d)
(K)
(K)
(g)
(g)
(K)
(K)
(h)
(h)
Figure
10.
(a)
H*wind
wind
speed
Ritaon
on2222September
September2005,
2005,atat
at1330
1330UTC;
UTC;(b)
(b)Retrieved
Retrievedwind
wind
Figure
10.(a)
(a)H*wind
H*windwind
windspeed
speedofofRita
September
2005,
1330
UTC;
(b)
Retrieved
wind
Figure
10.
speed
of
Rita
with
our
new
model
2005,
at
1155
UTC;
(c)
RSS
all-weather
wind
speed
speed
of
Rita
with
our
new
model
on22
September
2005,
at
1155
UTC;
(c)
RSS
all-weather
wind
speed
speed of Rita with our new model on22 September 2005, at 1155 UTC; (c) RSS all-weather wind
Rita
on
22
September
2005,
at 1155
UTC;
(d) RSS
of
Rita
22
atatUTC;
1155
Rita
September
2005,
at
rain
rate
of on
Rita
on
22 September
September
2005,
1155
ofofof
Rita
onon
2222
September
2005,
at 1155
UTC;
(d) RSS
rain rain
rate rate
of
Rita
22on
September
2005, 2005,
at
1155
UTC;
(e) The
The
bias between
between
H*wind
wind
speed
(f)
The
UTC;
bias
and Retrieved
wind
speed
with
our new
new
model;
(f)between
Thebias
bias
(e)
The (e)
bias
between
H*wind H*wind
and Retrieved
wind speed
with
ourwith
new our
model;
(f)model;
The bias
between
H*wind
and RSS
RSS all-weather
all-weather
W6H
over
on
1155
between
H*wind
and
wind
W6H
over
Rita
on22
22September
September
2005,
1155
H*wind
and
RSS all-weather
wind speed;
(g)speed;
W6H(g)
over
Rita
on Rita
22
September
2005, at2005,
1155atat
UTC;
UTC;
(h)
W6V
over
Rita
on 22
22 September
September
1155 UTC.
UTC;
(h)
W6V
over
on
UTC.
(h)
W6V
over
Rita
on Rita
22
September
2005, at2005,
1155 at
UTC.
Figure 11. (a) H*wind wind speed for Rita on 22 September 2005, at 0130 UTC; (b) Retrieved wind speed
Figure 11. (a) H*wind wind speed for Rita on 22 September 2005, at 0130 UTC; (b) Retrieved wind
Figure
H*wind
speed
Rita on 22
September
2005,(c)atRSS
0130
UTC; (b) wind
Retrieved
based
on11.
our(a)
new
model wind
for Rita
on 21for
September
2005,
at 2328 UTC;
all-weather
speedwind
for
speed based on our new model for Rita on 21 September 2005, at 2328 UTC; (c) RSS all-weather wind
speed
based
on our new
forUTC;
Rita (d)
on RSS
21 September
2005,
2328
(c) RSS
all-weather
wind
Rita
on 21
September
2005,model
at 2328
rain rate for
Ritaaton
21 UTC;
September
2005,
at 2328 UTC;
speed for Rita on 21 September 2005, at 2328 UTC; (d) RSS rain rate for Rita on 21 September 2005, at
speed
for Rita
on 21H*wind
September
at 2328
UTC;
(d) RSSwith
rain our
ratenew
for Rita
on 21
2005, at
(e)
The bias
between
and2005,
the wind
speed
retrieved
model;
(f) September
The bias between
2328 UTC; (e) The bias between H*wind and the wind speed retrieved with our new model; (f) The
2328 UTC;
betweenwind
H*wind
and(g)
theW6H
windover
speed
retrieved
with our 2005,
new model;
The
H*wind
and(e)
theThe
RSSbias
all-weather
speed;
Rita
on 21 September
at 2328 (f)
UTC;
bias between H*wind and the RSS all-weather wind speed; (g) W6H over Rita on 21 September 2005,
bias
between
H*wind
the RSS 2005,
all-weather
(h)
W6V
over Rita
on 21and
September
at 2328wind
UTC.speed; (g) W6H over Rita on 21 September 2005,
at 2328 UTC; (h) W6V over Rita on 21 September 2005, at 2328 UTC.
at 2328 UTC; (h) W6V over Rita on 21 September 2005, at 2328 UTC.
4.4.
Rain
Effects
4.4.
Rain
EffectsononWind
WindRetrieval
Retrieval
4.4. Rain Effects on Wind Retrieval
The
passive
radiometer
The passive radiometermeasurement
measurementisisless
lesssensitive
sensitiveto
tothe
theocean
ocean surface
surface wind
wind speed
speed under
under
The
passive
radiometer
measurement
is
less
sensitive
to
the
ocean
surface
wind
speed
under
rainy
conditions.
Rain
increases
the
atmospheric
attenuation,
which
is
obvious
at
higher
frequencies
rainy conditions. Rain increases the atmospheric attenuation, which is obvious at higher frequencies
rainy conditions.
Rain increasesdifficult
the atmospheric attenuation,
whichbrightness
is obvious at higher frequencies
(Figure
3).3).
InIn
addition,
in rain,
rain,
(Figure
addition,it itisisvery
very difficulttotoaccurately
accuratelysimulate
simulatethe
the brightness temperatures
temperatures in
(Figure
3).
In
addition,
it
is
very
difficult
to
accurately
simulate
the
brightness
temperatures
in
especially
for
the
high
variability
of
rainy
atmospheres.
The
brightness
temperatures
depend
on
cloud
especially for the high variability of rainy atmospheres. The brightness temperatures dependrain,
on
especially
for
the
high
variability
of
rainy
atmospheres.
The
brightness
temperatures
depend
on
type
and
the
distribution
of
rain
within
the
footprint
[32,51].
Then,
the
full
Mie
absorption
theory
cloud type and the distribution of rain within the footprint [32,51]. Then, the full Mie absorption
cloud
and
distribution
of rain the
within
the footprint
[32,51].
Then,
the full which
Mie absorption
theorytype
needs
to the
be applied
to calculate
atmospheric
radiative
transfer
equation,
requires
theory
needs
to
be
applied
to
calculate
the
atmospheric
radiative
transfer
equation,
which requires
additional input parameters, such as size and form of the rain drops; however, these parameters
are
additional
input
parameters,
such
as
size
and
form
of
the
rain
drops;
however,
these
parameters
are
not readily available. Therefore, previous studies have mainly focused on the statistical algorithm for
not
readily
available.
Therefore,
previous
studies
have
mainly
focused
on
the
statistical
algorithm
for
retrieving wind speed under rainy conditions [10,15,17].
retrieving wind speed under rainy conditions [10,15,17].
Remote Sens. 2016, 8, 721
17 of 20
needs to be applied to calculate the atmospheric radiative transfer equation, which requires additional
input parameters, such as size and form of the rain drops; however, these parameters are not readily
available. Therefore, previous studies have mainly focused on the statistical algorithm for retrieving
wind speed under rainy conditions [10,15,17].
Using the RSS rain rate products, the rain effect is evaluated in the new model. Table 4 presents
the biases and RMS errors of the difference between the wind speeds from the retrievals and the
corresponding H*wind analysis wind speeds as a function of the WindSat rain rate. In general,
the biases are negative when the rain rate is less than 4 mm/h or more than 14 mm/h. Therefore,
the retrieved wind speed underestimates wind speed in this situation. The biases become positive
when the rain rate ranges from 4 to 14 mm/h. In general, the wind speed retrieval errors increase with
increasing rain rate. The wind speed retrieval accuracy of our new model inside hurricane ranges from
2.0 m/s in light rain to 3.9 m/s in heavy rain.
Table 4. The statistics for different rain intervals between H*wind wind speed and the wind speed
retrieved with our new method.
Rain Interval (mm/h)
Average Rain Rate (mm/h)
Number
Bias (m/s)
RMS (m/s)
[0, 2]
[2, 4]
[4, 6]
[6, 8]
[8, 10]
[10, 12]
[12, 14]
above 14
0.55
3.02
5.03
7.00
9.06
10.95
12.86
14.73
10495
3241
3532
2920
2633
1869
1044
508
−0.37
−0.11
0.18
0.57
0.33
0.73
0.49
−0.57
1.96
2.52
2.72
2.85
3.15
3.59
3.73
3.89
5. Conclusions
In this study, we assess the wind-induced emissivity changes in C-band and X-band channels
using simulated brightness temperatures. Four parameters (6V*, 6H*, 10V*, and 10H*) are defined,
and they present the quantity of ocean microwave emission changed by ocean wind. The results show
that the sensitivity of 6(10)V* to wind speed is approximately 0.37(0.5) K/(m/s) and that the sensitivity
of 6(10)H* to wind speed is approximately 0.85(0.93) K/(m/s). Furthermore, we find that the ratio of
6V(H)* to 10V(H)* is a constant, which is approximately 0.82(0.92). These results are in good agreement
with previous studies [15,28]. Based on the work in [15], we define two new parameters, W6H and
W6V, which represent increments in 6H and 6V due to ocean wind, respectively. Finally, a wind speed
retrieval algorithm for wind speeds above 20 m/s in hurricanes has been developed using the two new
parameters. The C-band and X-band vertically and horizontally polarized brightness temperatures are
used to retrieve high wind speeds because the brightness temperature acquired at higher frequencies
become saturated inside hurricanes.
The WindSat measurements over 15 hurricanes from 2003 to 2010 and the corresponding H*wind
analysis data are used to develop and validate the retrieval model. The performance is encouraging,
with a mean bias of 0.04 m/s and an overall RMS difference of 2.75 m/s. In general, the retrieved
wind speeds have a good agreement with the H*wind analysis wind speeds. Two case studies are
performed, resulting in a mean bias and RMS difference of 0.79 m/s and 1.79 m/s for hurricane Rita-1
and 0.63 m/s and 2.38 m/s for hurricane Rita-2. Compared to the RSS global all-weather wind speeds,
our retrieved wind speeds is closer to the H*wind analysis data, and the RSS global all-weather wind
speeds underestimates the wind speed. In addition, the retrieved wind speeds are compared to the
SFMR wind speeds. Good agreement is found between them.
It is a challenging job to retrieve wind speed under rainy conditions, especially for typhoons and
hurricanes. The core of the new algorithm for determining the wind speed under rainy conditions
involves using different channel combinations that reduce the signal coming from rain without greatly
Remote Sens. 2016, 8, 721
18 of 20
reducing the wind speed signal. With respect to the effect of rain, the performance of the new model
inside hurricanes ranges from 2.0 m/s in light rain to 3.9 m/s in heavy rain. The performance of
retrieving wind speed under rainy conditions is satisfactory.
This new algorithm can be used to retrieve wind speeds in hurricanes under rainy conditions with
an encouraging degree of accuracy, although we cannot expect the algorithm to have the same accuracy
as in rain-free cases. It is noted that the retrieved wind speed always overestimates wind speed in
the eye of the hurricane. That may be because there are still strong waves that increase the emission
from the ocean surface, even if the wind speed is relatively low in the eye of hurricane. For practical
applications of the new wind speed algorithm, we should update the current method for hurricanes
in which the areas of high wind and high rain are not fully observed. In addition, the relationship
between wind speed and wind-induced emissivity under different sea states, wave ages, and air-sea
stability in hurricanes should be studied. Furthermore, the wind direction retrieval will be studied
using the polarimetric brightness temperatures in the future.
Acknowledgments: This work was funded by the National Natural Science Foundation of China
(Grant Nos. 61501433; 41576171). This work was also supported by CAS frontier science key research projects
(Grant No. QYZDB-SSW-SYS010). We thank the Computational Physics. Inc. for providing WindSat brightness
temperature and the Hwind Corporate Office for providing the H*wind analysis data. WindSat global all-weather
wind product and rain rates are produced by Remote Sensing Systems and sponsored by the NASA Earth Science
MEaSUREs DISCOVER Project and the NASA Earth Science Physical Oceanography Program. SFMR data are
provided by the Hurricane Research Division, NOAA.
Author Contributions: L.Z. and X.Y. originated the main concept of this study. L.Z. carried out the experiments,
analysis and writing in this study; X.Y. provided some constructive suggestions and revised the manuscript;
H.S. and Z.W. contributed to the discussion of the results.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Gentemann, C.L.; Wentz, F.J.; Brewer, M.; Hilburn, K.; Smith, D. Passive microwave remote sensing of the
ocean: An overview. In Oceanography from Space; Barale, V., Gower, J.F.R., Alberotanza, L., Eds.; Springer:
Hague, The Netherlands, 2010; pp. 13–33.
Hollinger, J.P.; Peirce, J.L.; Poe, G.A. SSM/I instrument evaluation. IEEE Trans. Geosci. Remote Sens. 1990, 28,
781–790. [CrossRef]
Goodberlet, M.A.; Swift, C.T.; Wilkerson, J.C. Ocean surface wind speed measurements of the Special Sensor
Microwave/Imager (SSM/I). IEEE Trans. Geosci. Remote Sens. 1990, 28, 823–828. [CrossRef]
Atlas, R.; Hoffman, R.N.; Bloom, S.C.; Jusem, J.C.; Ardizzone, J. A multiyear global surface wind velocity
dataset using SSM/I wind observations. Bull. Am. Meteorol. Soc. 1996, 77, 869–882. [CrossRef]
Yu, L.; Jin, X. Buoy perspective of a high-resolution global ocean vector wind analysis constructed from
passive radiometers and active scatterometers (1987–present). J. Geophys. Res. 2012. [CrossRef]
Wentz, F.J. Measurement of oceanic wind vector using satellite microwave radiometers. IEEE Trans. Geosci.
Remote Sens. 1992, 30, 960–972. [CrossRef]
Wentz, F.J.; Meissner, T. AMSR Ocean Algorithm, Algorithm Theoretical Basis Document; RSS Technical Report
121599A-1; Remote Sensing Systems: Santa Rosa, CA, USA, 2000.
Meissner, T.; Wentz, F.J. Ocean retrievals for WindSat: Radiative transfer model, algorithm, validation.
In Proceedings of the 9th Specialist Meeting on Microwave Radiometry and Remote Sensing, San Juan, PR,
USA, 28 February–3 March 2006; pp. 4761–4764.
Bettenhausen, M.H.; Smith, C.K.; Bevilacqua, R.M.; Wang, N.Y.; Gaiser, P.W.; Cox, S. A nonlinear optimization
algorithm for WindSat wind vector retrievals. IEEE Trans. Geosci. Remote Sens. 2006, 44, 597–610. [CrossRef]
Meissner, T.; Wentz, F.J. Wind-vector retrievals under rain with passive satellite microwave radiometers.
IEEE Trans. Geosci. Remote Sens. 2009, 47, 3065–3083. [CrossRef]
Neul, N.; Tenerelli, J.; Chapron, B.; Vandemark, D.; Quilfen, Y.; Kerr, Y. SMOS satellite L-band radiometer:
A new capability for ocean surface remote sensing in hurricanes. J. Geophys. Res. 2012. [CrossRef]
Yueh, S.H. Directional signals in Windsat observations of hurricane ocean winds. IEEE Trans. Geosci.
Remote Sens. 2008, 46, 130–136. [CrossRef]
Remote Sens. 2016, 8, 721
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
19 of 20
Yan, B.; Weng, F. Applications of AMSR-E measurements for tropical cyclone predictions part I: Retrieval of
sea surface temperature and wind speed. Adv. Atmos. Sci. 2008, 25, 227–245. [CrossRef]
Mai, M.; Zhang, B.; Li, X.F.; Hwang, P.A.; Zhang, J.A. Application of AMSR-E and AMSR2 Low-Frequency
Channel Brightness Temperature Data for Hurricane Wind Retrievals. IEEE Trans. Geosci. Remote Sens. 2016,
54, 4501–4512. [CrossRef]
Shibata, A. A wind speed retrieval algorithm by combining 6 and 10 GHz data from Advanced Microwave
Scanning Radiometer: Wind speed inside hurricanes. J. Oceanogr. 2006, 62, 351–359. [CrossRef]
El-Nimri, S.F.; Jones, W.L.; Uhlhorn, E.; Ruf, C.; Johnson, J.; Black, P. An improved C-band ocean surface
emissivity model at hurricane-force wind speeds over a wide range of earth incidence angles. IEEE Trans.
Geosci. Remote Lett. 2010, 7, 641–645. [CrossRef]
Zabolotskikh, E.V.; Mitnik, L.M.; Chapron, B. New approach for severe marine weather study using satellite
passive microwave sensing. Geophys. Res. Lett. 2013, 40, 3347–3350. [CrossRef]
Zabolotskikh, E.V.; Mitnik, L.M.; Chapron, B. GCOM-W1 AMSR2 and MetOp-A ASCAT wind speeds for the
extratropical cyclones over the North Atlantic. Remote Sens. Environ. 2014, 147, 89–98. [CrossRef]
Yao, P.; Wan, J.; Wang, J.; Zhang, J. Satellite retrieval of hurricane wind speeds using the AMSR2 microwave
radiometer. Chin. J. Oceanol. Limnol. 2015, 33, 1104–1114. [CrossRef]
Gaiser, P.W.; St. Germain, K.M.; Twarog, E.M.; Poe, G.A.; Purdy, W.; Richardson, D.; Grossman, W.;
Jones, W.L.; Spencer, D.; Golba, G.; et al. The WindSat spaceborne polarimetric microwave radiometer:
Sensor description and early orbit performance. IEEE Trans. Geosci. Remote Sens. 2004, 42, 2347–2361.
[CrossRef]
Yueh, S.H.; Stiles, B.W.; Tsai, W.Y.; Hu, H.; Liu, W.T. QuikSCAT geophysical model function for tropical
cyclones and application to Hurricane Floyd. IEEE Trans. Geosci. Remote Sens. 2001, 39, 2601–2612. [CrossRef]
Adams, I.S.; Hennon, C.C.; Jones, W.L.; Ahmad, K.A. Evaluation of hurricane ocean vector winds from
WindSat. IEEE Trans. Geosci. Remote Sens. 2006, 44, 656–667. [CrossRef]
Bessho, K.; DeMaria, M.; Knaff, J.A. Tropical cyclone wind retrievals from the Advanced Microwave
Sounding Unit: Application to surface wind analysis. J. Appl. Meteorol. Climatol. 2006, 45, 399–415.
[CrossRef]
Zhang, B.; Perrie, W. Recent progress on high wind-speed retrieval from multi-polarization SAR imagery:
A review. Int. J. Remote Sens. 2014, 35, 4031–4045. [CrossRef]
Powell, M.D.; Houston, S.H.; Amat, L.R.; Morisseau-Leroy, N. The HRD real-time hurricane wind analysis
system. J. Wind Eng. Ind. Aerod. 1998, 77, 53–64. [CrossRef]
Houston, S.H.; Shaffer, W.A.; Powell, M.D.; Chen, J. Comparisons of HRD and SLOSH surface wind fields in
hurricanes: Implications for storm surge modeling. Weather Forecast. 1999, 14, 671–686. [CrossRef]
Sampson, C.R.; Schrader, A.J. The automated tropical cyclone forecasting system (version 3.2). Bull. Am.
Meteorol. Soc. 2000, 81, 1231–1240. [CrossRef]
Uhlhorn, E.; Black, P. Verification of remotely sensed sea surface winds in hurricanes. J. Atmos. Ocean Technol.
2003, 140, 99–116. [CrossRef]
Uhlhorn, E.; Black, P.; Franklin, J.; Goodberlet, M.; Carswell, J.; Goldstein, A. Hurricane surface wind
measurements from an operational stepped frequency microwave radiometer. Mon. Weather Rev. 2007, 135,
3070–3085. [CrossRef]
Klotz, B.W.; Uhlhorn, E.W. Improved stepped frequency microwave radiometer tropical cyclone surface
winds in heavy precipitation. J. Atmos. Ocean Technol. 2014, 31, 2392–2408. [CrossRef]
Yueh, S.H.; Stiles, B.W.; Liu, W.T. QuikSCAT wind retrievals for tropical cyclones. IEEE Trans. Geosci.
Remote Sens. 2003, 41, 2616–2628. [CrossRef]
Hilburn, K.A.; Wentz, F.J. Intercalibrated passive microwave rain products from the unified microwave
ocean retrieval algorithm (UMORA). J. Appl. Meteorol. Climatol. 2008, 47, 778–794. [CrossRef]
Chu, J.H. Tropical Cyslone Forecasters’s Reference Guide. Reference document initially published as NRL
Report NRL/PU/7541-95-0012. October 1995. Available online: http://www.nrlmry.navy.mil/%7Echu/
index.html (accessed on 21 March 2016).
Tsang, L. Polarimetic passive microwave remote sensing of random discrete scatterers and rough surfaces.
J. Electromagn. Wave 1991, 5, 41–57.
Yueh, S.H.; Kwok, R. Electromagnetic fluctuations for anisotropic media and the generalized Kirchhoff’s law.
Radio Sci. 1993, 28, 471–480. [CrossRef]
Remote Sens. 2016, 8, 721
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
20 of 20
Smith, C.K.; Bettenhausen, M.; Gaiser, P.W. A statistical approach to WindSat ocean surface wind vector
retrieval. IEEE Trans. Geosci. Remote Lett. 2006, 3, 164–168. [CrossRef]
Li, L.; Gaiser, P.W.; Gao, B.C.; Bevilacqua, R.M.; Jackson, T.J.; Njoku, E.G.; Rudiger, C.; Calvet, J.C.; Bindlish, R.
WindSat global soil moisture retrieval and validation. IEEE Trans. Geosci. Remote Sens. 2010, 48, 2224–2241.
[CrossRef]
Johnson, J.T. An efficient two-scale model for the computation of thermal emission and atmospheric reflection
from the sea surface. IEEE Trans. Geosci. Remote Sens. 2006, 44, 560–568. [CrossRef]
Wentz, F.J. A model function for ocean microwave brightness temperatures. J. Geophys. Res. 1983, 88,
1892–1908. [CrossRef]
Shibata, A. A change of microwave radiation from the ocean surface induced by air-sea temperature
difference. Radio Sci. 2003. [CrossRef]
Ulaby, F.T.; Moore, R.K.; Fung, A.K. Microwave Remote Sensing Active and Passive-Volume III: From Theory to
Applications; Artech House: Dedham, MA, USA, 1986.
Chevallier, F. Sampled Databases of 60-Level Atmospheric Profiles from the ECMWF Analyses; Research Report
No.4; Eumetsat/ECMWF SAF Programme: Darmstadt, Germany, 2001.
Courtier, P.; Andersson, E.; Heckley, W.; Vasiljevic, D.; Hamrud, M.; Hollingsworth, A.; Rabier, F.;
Fisher, M.; Pailleux, J. The ECMWF implementation of three-dimensional variational assimilation (3D-Var).
I: Formulation. Q. J. R. Meteorol. Soc. 1998, 124, 1783–1807. [CrossRef]
Wentz, F.J. A well-calibrated ocean algorithm for special sensor microwave/imager. J. Geophys. Res. 1997,
102, 8703–8718. [CrossRef]
Liebe, H.J.; Hufford, G.A.; Cotton, M.G. Propagation modeling of moist air and suspended water/ice particles
at frequencies below 1000 GHz. In Proceedings of the AGARD 52nd Specialists’s Meeting Electromagnetic
Wave Propagation Panel, Palma de Mallorca, Spain, 17–21 May 1993.
Rosenkranz, P.W. Water vapor microwave continuum absorption: A comparison of measurements and
models. Radio Sci. 1998, 33, 919–928. [CrossRef]
Hocking, J.; Rayer, P.J.; Rundle, D.; Saunders, R.W.; Matricardi, M.; Geer, A.; Brunel, P.; Vidot, J. RTTOV v11
Users Guide; NWP-SAF Report; Met. Office: Exeter, UK, 2014.
Shibata, A. AMSR/AMSR-E SST algorithm developments—Removal of ocean wind effect. Italian J.
Remote Sens. 2004, 30, 131–142.
Sasaki, Y.; Asanuma, I.; Muneyama, K.; Naito, G.; Suzuki, T. Microwave emission and reflection from
the wind-roughened sea surface at 6.7 and 18.6 GHz. IEEE Trans. Geosci. Remote Sens. 1988, 26, 860–868.
[CrossRef]
Klein, L.; Swift, C. An improved model for the dielectric constant of sea water at microwave frequencies.
IEEE Trans. Antennas Propag. 1977, 25, 104–111. [CrossRef]
Wentz, F.J.; Spencer, R.W. SSM/I rain retrievals within a unified all-weather ocean algorithm. J. Atmos. Sci.
1998, 55, 1613–1627. [CrossRef]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).