IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 41, NO. 10, OCTOBER 2003 2307
Sea S a e E ec on he Sea Su ace
Emissi i y a L-Band
Jo ge José Mi anda, Me cè Vall-llosse a, Membe , IEEE, Ad iano Camps, Senio Membe , IEEE,
Nú ia Du o, Membe , IEEE, Ignasi Co bella, Membe , IEEE, and Jacqueline E che o
Abs ac —In May 1999, he Eu opean Space Agency (ESA)
selec ed he Ea h Explo e Oppo uni y Soil Mois u e and
Ocean Salini y (SMOS) mission o ob ain global and equen soil
mois u e and ocean salini y maps. SMOS single payload is he
Mic owa e Imaging Radiome e by Ape u e Syn hesis (MIRAS),
an L-band wo-dimensional ape u e syn hesis adiome e wi h
mul iangula obse a ion capabili ies. A L-band, he b igh ness
empe a u e sensi i i y o he sea su ace salini y (SSS) is low,
app oxima ely 0.5 K/psu a 20 C, dec easing o 0.25 K/psu a
0C, compa able o ha o he wind speed 0.2 K/(m/s) a
nadi . Howe e , a a gi en ime, he sea s a e does no depend only
on local winds, bu on he local wind his o y and he p esence
o wa es a eling om a dis ances. The Wnd and Salini y
Expe imen (WISE) 2000 and 2001 campaigns we e sponso ed by
ESA o de e mine he impac o oceanog aphic and a mosphe ic
a iables on he L-band b igh ness empe a u e a e ical and
ho izon al pola iza ions. This pape p esen s he esul s o he
analysis o h ee nons a iona y sea s a e condi ions: g owing and
dec easing sea, and he p esence o swell. Measu ed sea su ace
spec a a e compa ed wi h he heo e ical ones, compu ed using
he ins an aneous wind speed. Di e ences can be minimized using
an “e ec i e wind speed” ha makes he heo e ical spec um
bes ma ch he measu ed one. The impac on he p edic ed
b igh ness empe a u es is hen assessed using he small slope
app oxima ion/small pe u ba ion me hod (SSA/SPM).
Index Te ms—Nons a iona y sea condi ions, sea su ace emis-
si i y a L-band, sea su ace spec um, swell e ec .
I. INTRODUCTION
KNOWLEDGE o he dis ibu ion o sal in he ocean
and i s annual and in e annual a iabili y a e c ucial
in unde s anding he ole o he ocean in he clima e sys em.
The measu emen o he sea su ace salini y (SSS) is one
o he challenges o he Eu opean Space Agency’s (ESA)
Soil Mois u e and Ocean Salini y (SMOS) Ea h Explo e
Oppo uni y mission. SMOS single payload is he Mic owa e
Imaging Radiome e by Ape u e Syn hesis (MIRAS), a
wo-dimensional (2-D) L-band syn he ic ape u e adiome e
wi h ull-pola ime ic capabili y. Two-dimensional b igh ness
Manusc ip ecei ed Oc obe 14, 2002; e ised June 30, 2003. This wo k was
suppo ed by he Spanish Comisión In e minis e ial de Ciencia y Tecnología
(CICYT TIC99-1050-C03-01) and by G an “Dis inció de Rece ca de la Gen-
e ali a de Ca alunya,” esolu ion: UNI/2120/2002 on July 19 h, DOGC 3684,
24/07/2002. The WISE 2000 and 2001 ield campaigns ha e been sponso ed by
he Eu opean Space Agency unde ESTEC con ac 14188/00/NL/DC.
J. J. Mi anda, M. Vall-llosse a, A. Camps, N. Du o, and I. Co bella a e
wi h he Technical Uni e si y o Ca alonia (UPC), E-08034, Ba celona, Spain
(e-mail: [email p o ec ed]).
J. E che o is wi h Labo a oi e d’Océanog aphie Dynamique e Clima ologie
(LODYC), UPMC, F-75252 Pa is Cedex 05, F ance.
Digi al Objec Iden i ie 10.1109/TGRS.2003.817190
empe a u e images will p o ide looks o he same pixel unde
incidence angles om 0 o almos 65 , which equi es he
de elopmen o soil and sea emission models in he whole
ange o incidence angles, and sui able geophysical pa ame e s
e ie al algo i hms.
The dielec ic pe mi i i y o seawa e is de e mined, among
o he a iables, by salini y. The e o e, in p inciple, i is possible
o e ie e SSS om passi e mic owa e measu emen s as long
as he a iables in luencing he b igh ness empe a u e (TB)
signal (sea su ace empe a u e, oughness, and oam) can be
accoun ed o a di e en iewing angles, pola iza ions, and e-
quencies. The sensi i i y o TB o SSS is maximum a low mi-
c owa e equencies, and good condi ions o salini y e ie al
a e ound a L-band (1.4 GHz). Howe e , e en a his equency,
he sensi i i y o TB o SSS is s ill low ( 0.5 K/psu o a sea
su ace empe a u e SST 20 C, dec easing o 0.25 K/psu
o SST 0 C), which is compa able o ha o he wind speed
(0.2 K/(m/s) a nadi ). Consequen ly, i places demanding e-
qui emen s on bo h he pe o mance o he ins umen , and he
geophysical modeling o he sea su ace and i s emissi i y.
The SSS e ie al algo i hms o he SMOS mission mus
hen include: 1) he a ia ion o he local incidence angle, as a
pixel appea s in consecu i e snap-sho s, and e en ually 2) he
azimu hal a ia ion o he sea emissi i y o sea emission az-
imu hal signa u e. These wo poin s equi e an accu a e mod-
eling o he sea su ace emissi i y a L-band. Howe e , un il
ecen ly, expe imen al da a we e e y sca ce. The main goal
o he ESA-sponso ed Wind and Salini y Expe imen (WISE)
2000 and 2001 campaigns was o de e mine he TB sensi i i y
o wind speed (o sea s a e) [1], [2]. The campaigns ook place
a Repsol’s Casablanca oil ig,loca ed a 40 43.02 N 21.50 E,
40 km away om he Eb o Ri e mou h on he coas o Ta -
agona, Spain. The sea bed is 165 m, and he sea condi ions a e
ep esen a i eo heMedi e aneanshel /slope egionwi hpe i-
odic in luence o he Eb o Ri e esh wa e plume. WISE 2000
da a acquisi ion spanned om No embe 25 o Decembe 18,
2000 and om Janua y 8 o Janua y 15, 2001, and WISE 2001
om Oc obe 23 o No embe 22, 2001.
I is known ha he wind is he majo dis u bing o ce o he
sea su ace [3]. In WISE 2000 and 2001, pa o he disc epan-
cies be ween he b igh ness empe a u e da a and he b igh ness
empe a u e compu ed using ins an aneous local wind speed a e
a ibu ed owa e e lec ionsin heoil igs uc u e, hep esence
o wa es o igina ed by winds ha ha e blown a away, and sea
s a e condi ions di e en om hose o ully de eloped sea [4].
In Sec ion II, a desc ip ion o some sea su ace oughness
models and he in ol ed pa ame e s a e p esen ed. Then, in
0196-2892/03$17.00 © 2003 IEEE
2308 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 41, NO. 10, OCTOBER 2003
Fig. 1. Dependence o he uni ied omnidi ec ional spec um wi h he in e se-wa e-age ( he used alues o

a e 0.84, 1, 2, 3, 4, and 5). Wind speed is 15 m/s.
Sec ion III, he mos ep esen a i e clima ologic condi ions
occu ed du ing WISE campaigns a e analyzed. In ac , we
p esen an in e compa ison o measu ed spec a in WISE
campaigns wi h he ones compu ed by models when he
simul aneously measu ed wind speed is applied. Finally,
he di e ence be ween he b igh ness empe a u es o bo h
spec a (measu ed and heo e ical depending only wi h he
ins an aneous wind speed) is ob ained. I is demons a ed ha
his di e ence is compa able o he b igh ness empe a u e SSS
signa u e. The e o e, in o de o achie e he equi ed accu acy
in SSS e ie al, i is necessa y o know he ac ual sea s a e
condi ions and no only he ins an aneous wind speed.
II. MODELING THE SEA STATE
The s a is ical desc ip ion o he sea su ace is ep esen ed by
i s di ec ional spec um , de ined as he Fou ie ans-
o m o he au oco ela ion unc ion o he su ace heigh . Sea
s a e e ol es bo h in ime and space. The e olu ion o spec al
sea su ace componen s a wa enumbe , a eling in he di-
ec ion, along a dis ance , du ing a ime , ollows he ene gy
balance o anspo equa ion o deep-wa e [5]
(1)
whe e is he g oup speed. The igh -hand side o (1) adds he
wind ene gy inpu ( ), he dissipa ion ( ), and he ans e -
ence o ene gy among equency componen s due o nonlinea
in e ac ions ( ) [6].
Due o ic ional d ag in he ai –su ace in e ace, i s sho
capilla y wa es a ede eloped. Pa o he capilla y wa e ene gy
is hen ans e ed o longe wa eleng h g a i y wa es, and as
ime elapses he wa eleng h associa ed o he spec al peak and
o al ene gy inc ease. This p ocess con inues un il an equilib-
ium s a e, in which he inpu wind ene gy balances he ene gy
dissipa ion, is eached. In his s a e, he sea is said o be ully
de eloped. The ime necessa y o de elop g a i y wa es is hen
highe han ha o de elop capilla y wa es. When wind s ops,
sho wa es disappea apidly, while long wa es ake longe and
may las o se e al days, which allows hem o p opaga e o e
long dis ances and add o locally gene a ed wa es. This phe-
nomenon is usually known as swell, and i is conside ed o con-
ain wa e-pe iods highe han 10 s, app oxima ely [7].
Thedi ec ional spec um [(1)]is di idedin o heom-
nidi ec ional spec um and he angula sp eading unc ion
[(2)]
(2)
The sp eading unc ion, which is e en and eal, can be decom-
posed by a Fou ie se ies expansion
(3)
and i is usually app oxima ed wi h he wo i s e ms [8]
(4)
Many sea oughness spec um models ha e been de eloped.
A comp ehensi e desc ip ion o hem is p esen ed in [8]. How-
e e , mos o hem assume ully de eloped sea s a e condi ions,
a si ua ion ha equi es ha he wind blows a cons an speed
and di ec ion du ing a gi en ime and dis ance. Fully de eloped
sea spec a depend only on he wind measu ed a a e e ence
heigh . In his wo k, we use he wind speed measu ed a 10 m
(). The dis ance ha a cons an wind ec o co e s is called
e ch ( , in me e s), and i s dimensionless alue is ,
whe e is he wa enumbe , de ined as [9]. As
he wind speed inc eases, bo h e ch and wind ac ion ime mus
MIRANDA e al.: SEA STATE EFFECT ON THE SEA SURFACE EMISSIVITY AT L-BAND 2309
(a)
(b) (c)
Fig. 2. G owing sea case. (a) Wind speed e e ed o 10-m heigh du ing 24 h be o e he nominal measu emen ime. (b) Buoy-measu ed spec um (solid line),
uni ied spec um compu ed wi h he equi alen wind speed
U
=
6.77 m/s (do ed line), and uni ied spec um compu ed wi h he 3-h a e aged wind speed
measu ed a 10-m heigh
U
=
7.73 m/s (dashed line). (c) Di e ence o b igh ness empe a u e a e ical (do ed line) and ho izon al (dashed line) pola iza ions
compu ed wi h he SSA/SPM me hod using he heo e ical uni ied spec um calcula ed wi h
U
=
6.77 m/s minus he one ob ained wi h he measu ed wind speed
U
=
7.73 m/s. The e o is posi i e o he ho izon al pola iza ion and nega i e o he e ical pola iza ion in acco dance wi h a spec um o e es ima ion.
inc ease o each he condi ions o ully de eloped sea [3]. In
WISE expe imen s, because o he oil ig si ua ion (see he in-
oduc ion), in mos caseswhen he wind blows om hewes o
no hwes , he sea canno be conside ed ully de eloped. Dom-
inan winds in he Casablanca oil ig a e om he no h and
no hwes . Winds om he Sou h a e much a e .
F om he s udy o an ex ensi e da ase o de eloping sea
s a es, Donelan e al. [10] and Dobson [11] ob ained a depen-
dence o he long-wa e spec um wi h he dimensionless in-
e se-wa e-age pa ame e , whe e is he wa e
phase speed a he spec al peak. Taking in o accoun he ela-
ionship be ween he dimensionless ene gy wi h he e ch [12],
El ouhaily [8] de i ed he ollowing e ch o in e se-wa e-age
ela ionship:
(5)
whe e is he angle be ween he wind and he dominan wa es
a hespec alpeakandi s alueisalwaysclose oze o,and hen
. Seas a e said o be ully de eloped, ma u e and young
when has alues closed o 0.84, 1 and 2, espec i ely.
Fig. 1 shows he s ong dependence o he uni ied omnidi ec-
ional spec um, desc ibed in [8] (wind speed equal o 15 m/s)
wi h he in e se-wa e-age ( a ying om 0.84 o 5).
Sec ion III p esen s some plo s o measu ed wind speed and
spec a du ing WISE campaigns, showing ha in mos condi-
ions he sea s a e canno be conside ed ully de eloped. The
impac o sea s a e on he b igh ness empe a u es is e alua ed
using he small slope app oxima ion/small pe u ba ion me hod
(SSA/SPM) model [13], [14]. This model p o ides good ag ee-
men wi h measu ed and i s sensi i i y o wind speed.
III. SEA STATE IN NONSTATIONARY CONDITIONS:IMPACT ON
THE BRIGHTNESS TEMPERATURE AT L-BAND
Du ingWISEcampaignswindspeedsa 2.6- and69-mheigh
we e eco ded and e e ed o 10 m using he ollowing [15]:
cm/s (6)
2310 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 41, NO. 10, OCTOBER 2003
(a)
(b) (c)
Fig. 3. Dec easing sea case. (a) Wind speed e e ed o 10-m heigh du ing 24 h be o e he nominal measu emen ime. (b) Buoy-measu ed spec um (solid
line), uni ied spec um compu ed wi h he equi alen wind speed
U
=
7.45 m/s (do ed line) and uni ied spec um compu ed wi h he 3-h a e aged wind speed
measu ed a 10 m
U
=
5.21 m/s (dashed line). (c) Di e enceo b igh ness empe a u e a e ical (do ed line) and ho izon al (dashed line) pola iza ions compu ed
wi h he SSA/SPM me hod wi h he heo e ical uni ied spec um calcula ed wi h
U
=
7.45 m/s and
U
=
5.21 m/s. The e o is posi i e o he e ical
pola iza ion and nega i e o he ho izon al pola iza ion in acco dance wi h a spec um unde es ima ion.
whe e and a e he wind speed a heigh and he ic ion
speed, espec i ely, and is ob ained om
(7)
Fu he mo e, he sea su ace ele a ion spec a we e simul a-
neously measu ed by a wa e- ide buoy. Du ing 3 h, he buoy
measu es he sea heigh , and hen, i is Fou ie ans o med.
Ne e heless, he measu ed spec a a e de ined wi h only 14
poin s and he accu acy o he nume ical me hod depends on
he numbe o poin s used in he nume ical in eg a ion o hesea
su ace spec um.Hence, he ene gy o each measu ed spec um
has been ob ained, and hen, an equi alen wind speed ( ) has
been es ima ed in o de o compu e a uni ied wind d i en spec-
um wi h he same ene gy as he measu ed one. The com-
pu ed wi h he wind d i en spec um o he equi alen wind
speed ( ), a di e en ep esen a i e momen s du ing WISE
campaigns,we ecompa edwi h he compu edwi h hewind
d i en spec um applying he simul aneously measu ed wind
speed. In ac , in o de o be consis en wi h he 3-h measu e-
men s o he buoy, i has been used he a e age o wind speeds
eco ded du ing he3ho hebuoymeasu emen s ( ). Th ee
e en s a e p esen ed below as examples o h ee eal sea con-
di ions: g owing sea, dec easing sea and sea a ec ed by swell.
None o hei spec a is well cha ac e ized by ully de eloped
models.
The b igh ness empe a u es we e compu ed using he
second-o de SSA/SPM emissi i y model [16]. Johnson and
Zhang ecall a heo e ical de elopmen o his me hod in [13].
In ha pape , he physics o he emission p ocess p edic ed
by SSA/SPM was cla i ied. The use o he SSA/SPM up o
he second-o de ob ains an expansion in su ace slope, wi h
ze o-o de e ms ep oducing la su ace emission esul s,
i s -o de e ms iden ically ze o, and he second-o de e ms
p o iding he i s p edic ion o changes om la su ace
b igh ness. Second-o de e ms ake he o m o an in eg al
o a se o weigh ing unc ions o e he su ace di ec ional
spec um. P ope ies o a di ec ional spec um esul in no i s
ha monic a ia ion being ob ained; a hi d-o de SSA/SPM is
MIRANDA e al.: SEA STATE EFFECT ON THE SEA SURFACE EMISSIVITY AT L-BAND 2311
(a)
(b) (c)
Fig. 4. Swell case. (a) Wind speed e e ed o 10-m heigh du ing 24 h be o e he nominal measu emen ime. (b) Buoy-measu ed spec um (solid line), uni ied
spec um compu ed wi h he 3-h a e aged wind speed measu ed a 10-m heigh
U
=
11.33 m/s (do ed line) and he same spec um plus swell spec um wi h he
pa ame e sshownin hebox(dashedline).(c) Di e enceo b igh ness empe a u e a e ical(do edline)and ho izon al (dashedline)pola iza ionscompu ed wi h
he SSA/SPM me hod wi h he heo e ical uni ied spec um calcula ed wi h he equi alen wind speed and he a e aged measu ed wind speed. The swell inc eases
he oughness. The e o is posi i e o he e ical pola iza ion and nega i e o he ho izon al pola iza ion in acco dance wi h a spec um unde es ima ion.
(a) (b)
Fig. 5. (a) Signi ican wa e heigh (SWH) de ined as H1/3 he a e age o he highes hi d o he wa es [m], and (b) maximum wa e pe iod [s] om Oc obe 15
o No embe 22, 2001.

2312 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 41, NO. 10, OCTOBER 2003
(a) (b)
(c) (d)
(e) ( )
Fig. 6. B igh ness empe a u e di e ence be ween he uni ied heo e ical spec um wi hou swell and wi h swell. (a) and (c) Resul o he ho izon al pola iza ion.
(b) and (d) Resul o he e ical pola iza ion. In (a) and (b), he swell di ec ion ela i e o he wind di ec ion is 0 . In (c) and (d), he swell di ec ion is 45 espec
o he wind di ec ion. Plo (e) compa es he cu s a

=
30 in (a) and (c), and plo ( ) compa es he cu s a

=
30 in (b) and (d). The objec i e o (e) and ( ) is
o show he shi ing in he e o signa u e due o he angle be ween he wind and he swell di ec ions.
MIRANDA e al.: SEA STATE EFFECT ON THE SEA SURFACE EMISSIVITY AT L-BAND 2313
equi ed o ob ain he i s ha monics. Only he second-o de
expansion is conside ed in his s udy. No a i icial cu o
wa enumbe is equi ed o sepa a e small and la ge scale wa es
in his me hod.
A. G owing Sea
Fig. 2 co esponds o measu emen s held on No embe 1,
2001 a 6 A.M. (day o yea 305) and shows an example o a
g owing sea ha has no eached he ully de eloped condi ions.
The wind speed e olu ion du ing he p e ious 24 h is p esen ed
in Fig. 2(a), and shows wo wind speed jumps: he i s one 6 h
be o e he measu emen and he second one only hal an hou .
A la ge disc epancy be ween he measu ed spec um(solid line)
and he uni ied spec um o a ully de eloped sea (do ed line)
compu ed wi h he 3-h a e aged wind speed a 10 m om sea
le el, 7.73 m/s, is obse ed in Fig. 2(b). The measu ed
spec umcouldalsobeapp oxima edusinganin e se-wa e-age
la ge han 0.84 ( ully de eloped sea), which means ha he
sea has no eached ma u i y condi ions. Ne e heless, he dis-
ag eemen is s ill conside able. This is because he heo e ical
g owingspec umis he ansien s a ebe weena la seaand he
ully de eloped sea unde a cons an wind blowing o a qui e
long ime, whe eas heac ual si ua ion is hesum o noncons an
local wind e ec , plus he con ibu ion o he ene gy p opaga ed
om emo e places. An equi alen wind speed 6.27 m/s
has been ob ained. This wind speed has an associa ed uni ied
spec um [dashed line in Fig. 2(b)] wi h he same ene gy as
he measu ed one. Fig. 2(c) shows he di e ence be ween he
b igh ness empe a u es compu ed by SSA/SSPM o he ully
de eloped uni ied spec a wi h (same ene gy han measu e-
men ) and (a e age o 3-h measu ed wind speed). The e o
isnegligiblea nadi ,bu inc easeswi h ele a ionangle eaching
alues a ound 0.1 K and 0.12 K o and , espec i ely.
As compa ed o he salini y b igh ness empe a u e signa u e
(see he in oduc ion), hese alues a e no negligible a all.
B. Dec easing Sea
Fig. 3 co esponds o measu emen s on No embe 4, 2001 a
12 A.M. (day o yea 308) and ep esen s an example o a de-
c easing sea. The wind speed e olu ion du ing he p e ious 24 h
is plo ed in Fig. 3(a). The wind speed ends o dec ease om
20 h be o e he measu emen . Fig. 3(b) shows he measu ed
spec um (solid line), he uni ied spec um o a ully de eloped
sea o he 3-h a e aged wind speed a 10 m 5.21 m/s
(do ed line) and he uni ied spec um o a ully de eloped sea
conside ing he equi alen wind speed 7.45 m/s (dashed
line). As expec ed, he measu ed spec um shows a g ea e con-
ibu ion o he long wa es han he co esponding heo e ical
ully de eloped spec um. In gene al, he ine ia o he long
wa es c ea es a emnan oughness o e he sea su ace, which
p opaga es long dis ances and leads o he appea ance o swell
(example C).
Fig. 3(c) shows he di e ence be ween he b igh ness em-
pe a u es compu ed o he equi alen wind speed ( ) and he
a e aged wind speed ( ). The e o is negligible a nadi bu
inc eases wi h ele a ion angle eaching alues a ound 0.22 and
0.2 o and , espec i ely. This plo shows an opposi e
beha io wi h espec o he one in Fig. 2(c), which p esen s
he b igh ness empe a u e e o o an example o g owing sea.
Now, hee o isposi i e o heho izon alpola iza ionandneg-
a i e o he e ical one. In he p e ious case, he es ima ed
oughness was g ea e han he eal one, while o dec easing
sea he es ima ed oughness is lowe han he eal one. The en-
dency o he b igh ness empe a u e e o is consis en wi h he
emissi i y dependence o he wind, and hence, o he sea su -
ace oughness: an inc emen o he sea oughness p oduces an
inc emen o and a dec emen o .
C. Swell E ec s
The spec um is usually he esul o he local wind gene a ed
oughness and he swell p opaga ed om o he places. Plo s in
Fig. 4 co esponds o measu emen s on No embe 10, 2001 a
9P.M. day o yea 314) and show an example o he swell
e ec on he spec um and i s emissi i y. The wind speed e o-
lu ion du ing he p e ious 24 h is mo e cons an han in he
p e ious wo examples, bu in p e ious days [Fig. 5(a)] in ense
wind blew and signi ican wa e heigh s (de ined as he a e age
o one hi d o he highes wa es) o abou 1.8 m we e s ill
p esen in he a ea [Fig. 5(b)].
Swell spec um can be modeled by a na owband Gaussian
p ocess wi h a 2-D oughness spec um [17] wi h pa ame e s:
swell heigh a iance , di ec ional s anda d de ia ions
and , and swell spec al peak wa e numbe s and ,
in he di ec ions and , espec i ely, which a e no usually
coinciden wi h he up-wind and c oss-wind di ec ions
(8)
Then, he con ibu ion o he swell o he omnidi ec ional
spec um can be ob ained using he ollowing ela ion:
(9)
Fig. 4(b) shows he omnidi ec ional sea su ace spec um
measu ed by he wa e ide buoy (solid line), he uni ied
ully de eloped spec um o 3-h a e aged wind speed
11.33 m/s (dashed line), and his spec um plus a swell
spec um . The swell con ibu ion o he spec um has
been ob ained wi h he ollowing pa ame e s: 0.28 m,
0.011 m , and 0.043 ad/m
(do ed line). These alues ha e been chosen in o de o
minimize he di e ence be ween he measu ed spec um and
he wind-d i en ully de eloped uni ied spec um. I is clea
ha he combined model (do ed line) shows a be e ag eemen
wi h he measu ed spec um han jus he ully de eloped sea
spec um.
Fig. 4(c) shows he di e ence be ween he compu ed b igh -
ness empe a u es a e ical and ho izon al pola iza ions (
and ) when adding he swell e ec o no . As in he p e-
iouscase, he o al oughness(swellandwinde ec s)isg ea e
han he heo e ical one. The e o e, he e o o he e ical po-
la iza ion is posi i e, whe eas i is nega i e o he ho izon al
pola iza ion.
2314 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 41, NO. 10, OCTOBER 2003
Finally, since he swell has a e y di ec ional beha io ,
in Fig. 6 he same di e ence (emissi i y o a sea su ace
oughened by swell and wind and he one oughened only by
wind) is plo ed espec o azimu h angle as well. The azimu hal
signa u e is qui e small and dec eases wi h he ele a ion angle.
Fig. 6(a) and (b) shows he dependence o he e o when he
swell e ec is neglec ed o he e ical and he ho izon al
pola iza ions, espec i ely. In hose plo s he swell and wind
di ec ions a e he same. I he angle be ween he swell and he
wind di ec ions is 45 , his e o is only shi ed [see Fig. 6(c)
and (d)]. In o de o be e app ecia e his shi ing, Fig. 6(e)
and ( ) plo s he 30 cu in hose igu es. A shi o 45
is app ecia ed be ween do ed lines (swell and wind wi h he
same di ec ion) and solid line (45 be ween he swell and he
wind di ec ions).
IV. CONCLUSION
This pape demons a es ha i is impo an o know he ac-
ual sea s a e in he p edic ion o he L-band adiome ic emis-
sion. Fi s o all, ully de eloped wind-d i en uni ied spec a
compu ed wi h he 3-h a e aged measu ed wind a 10 m
ha e been compa ed wi h he measu ed spec a du ing WISE
campaigns, and a g ea disag eemen has been app ecia ed. The
measu ed spec a a e de ined wi h only 14 poin s and he accu-
acy o he emissi i y me hod depends on he numbe o poin s
used in he nume ical in eg al. Hence, he ene gy o each mea-
su ed spec um has been calcula ed, and hen, an equi alen
wind speed a 10 m ( ) has been es ima ed in o de o ob-
ain a ully de eloped wind-d i en uni ied spec um wi h he
same ene gy as he measu ed one. Finally, he di e ence o he
emissi i y using he spec um and he spec um, o
he ho izon al and he e ical pola iza ions has been compu ed
and plo ed wi h espec o he incidence angle. Th ee di e en
si ua ions ha e been analyzed: g owing sea, dec easing sea and
sea a ec ed by swell. In all o hem, he modulus o he di e -
ence ends o inc ease wi h he ele a ion angle. Di e ences o
a ac ion o a kel in a e ound, which a e no negligible when
e ie ing salini y om TB. These di e ences a e in he same
o de o magni ude as he dependence o TB wi h espec o he
sea su ace salini y. In addi ion i has been obse ed ha hese
di e ences ha e opposi e sign o he e ical and he ho izon al
pola iza ions: when he oughness is o e es ima ed is pos-
i i e and is nega i e, whe eas, when he oughness is un-
de es ima ed he beha io is jus he con a y. This is consis en
wi h he emissi i y dependence o he wind speed, and hence o
he sea su ace oughness: i he sea su ace oughness inc eases
he inc easesand he dec eases.The e o e,becauseo his
con a y beha io , his e o can be minimized by using he i s
S okes pa ame e in he SSS e ie al algo i hm [18].
The azimu hal dependence o he b igh ness empe a u e
e o when he swell e ec in he spec um is neglec ed has
also been s udied. This azimu hal dependence is always e y
small and dec eases wi h he ele a ion angle. Azimu h angles
wi h maximum e o a e shi ed, i he angle be ween he wind
and he swell di ec ions changes.
ACKNOWLEDGMENT
The au ho s app ecia e he coope a ion p o ided by he pe -
sonnel o Repsol In es igaciones Pe olí e as-Base Ta agona-
Pla a o ma Casablanca du ing WISE 2000 and 2001 cam-
paigns. The wa e- ide buoy measu emen s we e p o ided by
he Labo a oi e d’Océanog aphie Dynamique e Clima ologie
(LODYC).
REFERENCES
[1] A. Camps, J. Fon , J. E che o, V. Caselles, A. Weill, I. Co bella, M.
Vall-llosse a, N. Du o, F. To es, R. Villa ino, L. En ique, A. Julià, C.
Gaba ó, J. Bou in, E. Rubio, S. C. Reising, P. Wu s eisen, M. Be ge ,
and M. Ma ín-Nei a, “Sea su ace emissi i y obse a ions a L-band:
Fi s esul s o he wind and salini y expe imen WISE-2000,” IEEE
T ans. Geosci. Remo e Sensing, ol. 40, pp. 2117–2130, Oc . 2002.
[2] A. Camps, J. Fon , J. E che o, A. Weill, V. Caselles, I. Co bella, M.
Vall-llosse a, F.To es, N.Du o,R.Villa ino,L.En ique,J.Mi anda,A.
Julià, C. Gaba ó, J. Bou in, R. Niclós, P. Wu s eisen, M. Be ge , and M.
Ma ín-Nei a, “L-band sea su ace emissi i y adiome ic obse a ions
unde high winds: P elimina y esul s o he wind and salini y expe i-
men WISE-2001,” in P oc. IGARSS, 2002.
[3] J. A. Knauss, In oduc ion o Physical Oceanog aphy. Uppe Saddle
Ri e , NJ: P en ice-Hall, 1997.
[4] J. Mi anda, M. Vall-llosse a, A. Camps, and N. Du o, “Sea su ace
emissi i y a L-band: Swell e ec s expe imen WISE-2001,” in P oc.
IGARSS, 2002.
[5] K. Hasselmann, “On he nonlinea ene gy ans e in a g a i y-wa e
spec um pa gene al heo y,” J. Fluid Mech., ol. 15, no. 2, pp.
273–281, 1961.
[6] R. H. S ewa , In oduc ion o Physical Oceanog aphy. College S a-
ion, TX: Texas A&M Uni ., 2001.
[7] B. Kinsman, Wind Wa es. Thei Gene a ion and P opaga ion on he
Ocean Su ace. Mineola, NY: Do e , 1984.
[8] T. El ouhaily, B. Chap on, K. Ka sa os, and D. Vandema k, “A uni ied
di ec ional spec um o long and sho wind-d i en wa es,” J. Geophys.
Res., ol. 102, no. C7, pp. 15781–15796, 1997.
[9] S.A.Ki aigo odskii,“Applica ionso he heo yo simila i y o heanal-
ysis o wind-gene a ed wa e mo ion as a s ochas ic p ocess,” IZV. Akad.
Nauk SSSR, se . Se . Geo iz, ol. 1, p. 73, 1962.
[10] M. A. Donelan, J. Hamil on, and W. H. Hui, “Di ec ional spec a o
wind gene a ed wa es,” Philos. T ans. R. Soc. London A, ol. 315, pp.
509–562, 1985.
[11] F. Dobson, W. Pe ie, and B. Toulany, “On he deep-wa e e ch laws
o wind-gene a ed su ace g a i y wa es,” A mos. Ocean, ol. 27, no.
1, pp. 210–236, 1989.
[12] I. R. Young, L. A. Ve hagen, and M. L. Banne , “The e olu ion o wind
gene a edwa esin wa e o ini edep h,”in P oc.2ndIn .Con .Ai –Sea
In e ac ion and Me eo ology and Oceanog aphy o he Coas al Zone.
Lisbon, Po ugal, 1994.
[13] J. T. Johnson and M. Zhang, “Theo e ical s udy o he small slope
app oxima ion o ocean pola ime ic he mal emission,” IEEE T ans.
Geosci. Remo e Sensing, ol. 37, pp. 2305–2316, Sep . 1999.
[14] N. Reul and B. Chap on, “SMOS—salini y da a p ocessing s udy, im-
p o emen s in emissi i y models,” ESA, Noo dwijk, The Ne he lands,
Rep. WP 1100, 2001.
[15] F. T. Ulaby, R. K. Moo e, and A. K. Fung, Mic owa e Remo e Sensing
Ac i e and Passi e. No wood, MA: A ech House, 1990, ol. III, pp.
1560–1560.
[16] A. G. Vo ono ich, “Small-slope app oxima ion o elec omagne ic
wa e sca e ing a ough in e ace o wo dielec ic hal -spaces,” Wa es
Random Media, ol. 4, pp. 337–367, 1994.
[17] S. L. Du den and J. F. Vesecky, “A physical ada c oss-sec ion model
o a wind-d i en sea wi h swell,” IEEE J. Oceanic Eng., ol. OE-10,
pp. 445–451, Oc . 1985.
[18] A.Camps, I. Co bella, M.Vall-llosse a, N. Du o,F. To es, R. Villa ino,
L. En ique, F. Julbé, J. Fon , A. Julià, C. Gaba ó, J. E che o, J. Bou in,
A. Weill, V. Caselles, E. Rubio, P. Wu s eisen, and M. Ma ín-Nei a,
“L-Band sea su ace emissi i y: P elimina y esul s o he WISE-2000
campaign and i s applica ion o salini y e ie al in he SMOS mission,”
Radio Sci., ol. 38, no. 4, 2003.
MIRANDA e al.: SEA STATE EFFECT ON THE SEA SURFACE EMISSIVITY AT L-BAND 2315
Jo geJoséMi andawasbo n inLas PalmasdeG an
Cana ia, Cana y Islands, Spain, in 1973. He ecei ed
he Telecommunica ion Technical Enginee ing
deg ee om Las Palmas de G an Cana ia Uni e si y
(ULPGC), Spain, and he Senio Telecommunica ion
Enginee ing deg ee om he Poly echnic Uni e si y
o Ca alonia (UPC), Ba celona, Spain, in 1996 and
2001, espec i ely. He is cu en ly pu suing he
Ph.D. deg ee in elecommunica ion enginee ing
om UPC.
F om 1996 o 1999, he was wi h he Remo e
Sensing and Rada Labo a o y (EUITT-ULPGC), whe e collabo a ed
in sa elli e image p ocessing. In 1999, he joined he Signal Theo y and
Communica ions Depa men , UPC. His Ph.D. hesis is ocused on he sea
su ace emissi i y a L-band. I includes he de elopmen o new models o
de e mining he sea su ace oughness and he compa ison and imp o emen
o exis ing emissi i y models. Since 2002, he has been an Assis an P o esso
a he UPC.
Me cè Vall-llosse a (M’99) ecei ed he Senio Telecommunica ion Enginee
and he Doc o Telecommunica ion Enginee ing deg ees in 1990 and 1994, e-
spec i ely, bo h om he Poly echnicUni e si yo Ca alonia(UPC), Ba celona,
Spain.
Shehas been lec u ing and doing esea cha he Depa men o SignalTheo y
and Communica ions, UPC om 1990 un il 1997 as an Assis an P o esso and
om 1997 un il p esen as an Associa e P o esso . She spen a sabba ical yea
in Mon eal wi h he schola ship o he “P og amme Québécois de Bou ses
d’excellence”(1996-1997):“S agesdeFo ma ionpos doc o aleauQuébecpou
jeunes diplômés é ange s.” He esea ch in e es s include nume ical me hods
in elec omagne ism, mic owa e adiome y, an enna analysis, and design. Cu -
en ly, he esea ch is mainly ela ed o he s udy o nume ical me hods ap-
plied o he sea su ace emissi i y and hei cha ac e iza ion a L-band and he
MIRAS/SMOS p ojec .
D . Vall-llosse a, along wi h he o he membe o he adiome y g oup a
UPC, was awa ded he “P ime P emio Du an Fa ell de In es igación Tec-
nológica” in 2002, and he “P ime P emio Ciu a de Ba celona d’In es igació
Tecnològica” in 2001.
Ad iano Camps (S’91–A’97–M’00–SM’02) was
bo n in Ba celona, Spain, in 1969. He ecei ed he
Telecommunica ions Enginee ing deg ee and he
Ph.D. deg ee in elecommunica ions enginee ing in
1992 and 1996, espec i ely, bo h om he Poly-
echnic Uni e si y o Ca alonia (UPC), Ba celona,
Spain.
F om 1991 o 1992, he was wi h he ENS des Télé-
communica ions de B e agne, B e agne, F ance, wi h
an E asmus Fellowship. In 1993, he joined he Elec-
omagne icsandPho onicsEnginee ingg oup,a he
Depa men o Signal Theo y and Communica ions, UPC, as an Assis an P o-
esso , and since 1997 as an Associa e P o esso . In 1999, he was on sabba ical
lea e a he Mic owa e Remo e Sensing Labo a o y, Uni e si y o Massachu-
se s, Amhe s . His esea ch in e es s a e mic owa e emo e sensing, wi h spe-
cial emphasis in mic owa e adiome y by ape u e syn hesis echniques. He
has pe o med nume ous s udies wi hin he ame o Eu opean Space Agency
SMOS Ea h Explo e Mission. He is an Associa e Edi o o Radio Science.
D . Camps ecei ed he second na ional awa d o uni e si y s udies in
1993, he INDRA awa d o he Spanish Associa ion o Telecommunica ion
Enginee ing o he bes Ph.D. in 1997, he ex ao dina y Ph.D. awa d a he
Uni e si a Poli ècnica de Ca alunya in 1999, he Fi s Du an Fa ell Awa d
and he Ciudad de Ba celona Awa d, in 2000 and 2001, espec i ely, bo h o
Technology T ans e ; and in 2002, he Resea ch Dis inc ion o he Gene ali a
de Ca alunya o con ibu ions o mic owa e passi e emo e sensing. He was
Chai o Cal ’01. He is edi o o he IEEE Geoscience and Remo e Sensing
Newsle e and P esiden -Founde o he IEEE Geoscience and Remo e Sensing
Socie y Chap e a Spain.
Nú ia Du o (S’91–M’99) ecei ed he Telecom-
munica ion Enginee deg ee om he Poly echnic
Uni e si y o Ca alonia (UPC), Ba celona, Spain,
and he Doc o in Telecommunica ion Enginee ing
om UPC, in 1990 and 1996, espec i ely.
Since 1997, she has been an Associa e P o esso
a UPC. He cu en esea ch in e es s a e nume ical
me hods in elec omagne ics, mic owa e adiome y,
an enna analysis, and design.
Ignasi Co bella (M’99) ecei ed he Telecom-
munica ions Enginee ing and Doc o Enginee ing
deg ees, bo h om Uni e si a Poli ècnica de
Ca alunya (UPC), Ba celona, Spain, in 1977 and
1983, espec i ely.
In 1976, he joined he School o Telecommuni-
ca ion Enginee ing in UPC as a Resea ch Assis an
in he Mic owa e Labo a o y, we e he wo ked
on passi e mic owa e in eg a ed ci cui design
and cha ac e iza ion. Du ing 1979, he wo ked
a Thomson-CSF, Pa is, F ance, on mic owa e
oscilla o s design. In 1982, he became an Assis an P o esso a UPC, an
Associa e P o esso in 1986, and a Full P o esso in 1993. He is cu en ly
eaching mic owa es a he unde g adua e le el in UPC and has designed and
augh g adua e cou ses on nonlinea mic owa e ci cui s. Du ing he school
yea 1998–1999, he wo ked a NOAA/En i onmen al Technology Labo a o y,
Boulde , CO, as a Gues Resea che , de eloping me hods o adiome e
calib a ion and da a analysis. His esea ch wo k in he Depa men o Signal
Theo y and Communica ions, UPC includes mic owa e ai bo ne and sa elli e
adiome y and mic owa e sys em design.
Jacqueline E che o was bo n in Meknes, Mo occo,
in 1943. She ecei ed he Ag ega ion de Physique de-
g ee and he Doc eu es Sciences Physiques deg ee
o he hesis in space plasma physics in 1965 and
1972, espec i ely, bo h om he Ecole No male Su-
pe ieu e de Jeunes Filles, Pa is, F ance.
She became a Full-Time Scien is in Cen e Na-
ional de la Reche che Scien i ique in 1965. She is
p esen ly appoin ed o Labo a oi e d’Oceanog aphie
Dynamique e de Clima ologie, Pa is, F ance, as
Di ec eu de Reche ches. F om 1965 o 1987, she has
been wo king on space plasma physics, ocusing on s udies o p opaga ion and
gene a ion o wa es in he magne osphe e. She was esponsible o elaxa ion
sounde s onboa d GEOS1 (ESA, 1977), GEOS2 (ESA, 1978), and ISEE1
(ESA/NASA, 1977), including ins umen de ini ion, p o ision o ha dwa e,
da a handling, and scien i ic use o da a. In 1988, she became in e es ed in
oceanog aphy, using he syne gy be ween emo ely sensed measu emen s and
in si u da a o s udy he ai –sea CO2 exchange a la ge scale. He esea ch
ac i i y is mul iyea es ima ion o ai –sea CO2 lux a egional scale using
bo h sa elli e (wind speed, SST, ocean colo ) and in si u (PCO2 and ela ed
pa ame e s, including launching CARIOCA buoys) da a. Since 1999, she is
in ol ed in he p epa a ion o he SMOS sa elli e (ESA, 2007) in ended a
de e mining sea su ace salini y using L-band adiome y. She is in e es ed in
he de elopmen and alida ion o sea su ace emissi i y models a L-band,
including alida ion campaigns. She is coo dina ing he g oup o F ench
scien is s in ol ed in he ocean pa o he SMOS p ojec . She has published
77 pape s in he open li e a u e.