Ocean Salini y Obse a ions wi h SMOS Mission
M. Ma in-Nei a"'
,
J.
Fon o',
M.
S okosz@',
I.
C~ bella'~',
A.
Camps"'
Eu opean Space Agency
ESTEC, Keple laan 1,2200-AG Noo dwijk, The Ne he lands
Phone:
+31-71-565 4052; Fax:+31-71-565 4596;
E-mail: [email p o ec ed]
@)
Ins i u o de Ciencias del Ma
-
ICWCSIC (Spain)
"'Sou hamp on Oceanog aphy Cen m
-
SOC
(UK)
(4'Poly echnic Uni e si y
o
Ca alonia
-
UPC (Spain)
E-mail: [email p o ec ed]; [email p o ec ed]; [email p o ec ed]; [email p o ec ed]
INTRODUCTION
The pu pose o
his
pape
is
o p esen he capabili ies o
SMOS (Soil Mois u e
and
Ocean
Salini y mission) o
he
global mapping o
ocean
salini y om space. SMOS
has
been selec ed by he Eu opean
Space
Agency
as
he second
Ea h Explo e Oppo uni y wi h a launch da e
in
June
2005. The senso emba ked
on
SMOS is
MIRAS,
a
Mic owa e Imaging Radiome e
wi h
Ape u e Syn hesis.
MIRAS
wo ks a L-band, in he wo-pola isa ions, and
has
ull pola ime ic capabili y.
The measu emen o
sea
su ace
salini y
(SSS)
is one o
he challenges o
SMOS.
This
pape p esen s
i s
he
scien i ic equi emen s o a numbe o oceanog aphic
applica ions. The scien i ic equi emen s a e hen ansla ed
in o ins umen accu acy, sensi i i y, s abili y and spa ial
esolu ion Majo
sou ces
o e o in he e ie al o ocean
salini y will
be
add essed h ough
an
expe imen al
campaign
which is desc ibed.
SCIENTIFIC OBJECTIVES
Knowledge o he dis ibu ion o sal in he global
ocean
and i s
annual
and in e -annual a iabili y, a e c ucial in
unde s auding he ole o he
ocean
in
he clima e sys em.
Ocean
ci cula ion is mainly d i en by he momen um and
hea luxes h ough he a mosphe e-ocean in e ace, which
can
be
aced
by obse a ion o
SSS.
In addi ion, salini y
also de e mines
ocean
densi y and hence he mohaline
ci cula ion.
In
some egions (e.g. he A c ic), salini y is he
mos impo an a iable as i con ols p ocesses such as
deep wa e o ma ion which is a key componen in he
ocean
he mohaline ci cula ion ((con eyo belb). Ocean
salini y is also linked o he oceanic
ca bon
cycle, as i
plays a pa in es ablishing he chemical equilib ium, which
in
u n
egula es he CO2 up ake and elease. The e o e, he
assimila ion o
SSS
in o global
ocean
biogeochemical
models could imp o e es ima es o he abso p ion o
CO2
by he
oceans.
Moni o ing
SSS
could also be used o imp o e he
quali y o ENSO (El
Niiio
-
Sou hem Oscilla ion)
p edic ion by nume ical models. P esen ly he models
assimila e empe a u e andlo al ime e de i ed
sea
le el
da a
only. The lack o salini y nxasu emen s esul s in
majo disc epancies be ween modeled and obse ed &ice
cu en s.
Fo example, a 0.5 psu i@ ac ical salini y
uni s)
e o accoun s o a
3.8
cm/s e o
in
geos ophic elociw a
1
km
dep h calcula ed om he co esponding su ;ice
alue.
This
is paaicula ly impo an
in
he Wes em
Equa o ial Paci ic whe e he e
is
a s ong ENSO- ela ed
nea -su ace salini y signal
and
whe e
zonal
ad ec ion is o
majo impo ance o ENS0 mechnisms.
SSS
is
co ela ed wi h es ima es o he ne e apo a ion
minus
p ecipi a ion
(E-P)
balance.
(E-P)
is di icul o
measu e accu a ely o e he
ocean,
so
global maps o
SSS
would p o ide a cons ain
on
es ima es o (E-P)
on
he
global
scale.
This
would gi e insigh s in o he phenomena
d i ing he he mohaline Ci culi3 iOn and
also
allow
alida ing la en hea
lux
es ima es The wa e
lux
h ough
he
sea
su ace laye s is c i ical o densi y s a i ica ion
and
s ongly in luences he mixed laye dep h
and
he in ensi y
o
su ace
cu en s.
The dielec ic cons au o seawa e
is
de e mined,
among o he a iables, by salini y. In p inciple i is possilde
o e ie e
SSS
om mic owa e mieasu emen s
as
long
as
a iables in luencing he b igh ness empe a u e
(TB)
signal
(SST, oughness, oam,) can be accoun ed o e.g. by lhe
use o di e en iewing angles, pola isa ions
and
equencies. The sensi i i y o TB o
SSS
is
maximum
a
low mic owa e equencies and he good condi ions o
salini y e ie al a e ound a L-band (1.4
GHz,
21cm
wa eleng h). Howe e , i mus be s essed ha a
his
equency he sensi i i y
o
TB o
SISS
is low
(0.5K
pe
psu
o
an
SST o 2O9C, dec easing o 0.25K pe psu o
an
SST
o
OO),
placing demanding equi emlen s
on
he pe o mance
o he ins umen ( e e o Fig.
1).
Due o
his
low sensi i i y,
and
he spa ial esolu ion
ha
can
be expec ed wi h a spacebo ne mic owa e
in e e ome ic adiome e , we a e
!s ill
no in a posi ion o
ob ain
SSS
da a
o mesoscale o egonal
s udies.
Howe e , se e al phenomena ex eimely ele an o la ge-
0-7803-6359-O/OO/$lO.OO
0
2000
IEEE
2552
scale and clima ic s udies can bene i om such an
obse a ion app oach: ba ie laye e ec s
on
opical
Pacl ic hea
lux,
halos e ic adjus men o hea s o age om
sea le el,
No h
A lan ic he mohaline ci cula ion, su ace
eshwa e
lux
balance, e c. Conside ing he esolu ion
cons ains he o e all goal o
SSS
e ie al om
SMOS
da a
is
o mee he GODAE (Global
Ocean
Da a
Assimila ion Expe imen ) op imised equi emen o open
ocean
SSS:
a
esolu ion o 0.1 psu o e 200
km
boxes e e y
10 days.
The emissi i i y o he sea su ace
is
a
unc ion
o
salini y, empe a u e and wind speed as a mino pa ame e .
The bes sensi i i y o he OS is ob ained a 1.4 GHz,
howe e , he in luence o
sea
empe a u e
can
no be
neglec ed e en a
hese
equencies. Hence, he
measu emen o
he
ocean salini y equi es he use o
ano he spacec a , which allows he e ie al o SST.
MIRAS
CONFIGURATION
The ins umen o
SMOS
mission is
MIRAS,
a
2D
in e e ome ic adiome e designed o mee he
main
equi emen s o he mission [l]. I consis s o a o al
o
81
small
an enna ecei e s uni o mly
a anged
on
a Y-shape
s uc u e, and
a
o al powe adiome e in i s cen e o
absolu e calib a ion. To a oid
alias,
he
minimum
an enna
spacing o
his
con igu a ion
is
hld3=0.57h, being
h
he
wa eleng h a
he
cen al
equency. Due o
he
high
b igh ness empe a u e con as be ween
he
Ea h and he
sky,
a limi ed amoun o aliasing is pe mi ed, as long
as
only a ec s he
sky,
so
he selec ed an enna spacing is se
o 0.89h. Fo 27 an ennas pe
a m,
his
gi es a o al
a m
leng h o abou 5.2
m,
co esponding o an angula
esolu ion o less
han
2’. The ins umen will be pu
on
a
low Ea h sun-synch onous o bi wi h he an enna bo esigh
il ed wi h espec o nadi ,
so
he
angula esolu ion
ansla es o a g ound
spa ial
esolu ion be e
han
50
Km.
1.41
GHz
B ghimss
Tempe a u e
s
T,S
a $
midmm
angle
1O2[-,..pu.u~c
--_
.-
3
88
i
_._.
I
I I
25
30
35
40
Mimi y
(€4
8L-
Figu e
1
B igh ness Tempe a u e s
SSS
o Di e en SST
(cou esy om
Ga y
Lage loe )
“E9
DOMAIN
111
Figu e
2
Typical MIRAS Field o View (cou esy CASA and GMV)
The on-g ound ield o iew
(FOV)
o he ins umen
has
a
dis o ed hexagon
shape
wi h cu ed sides, ha ing
each pixel di e en sizes and incidence angles (Fig.2). A
snapsho b igh ness empe a u e map o
his
ield o iew
is ob ained each 0.78
s,
wi h an a e age esolu ion o abou
5K
o e 200K [2]. Indi idual esolu ion o
a
gi en pixel
is
in
ac weigh ed by he an enna pa e n.
Due o he mo ion o he pla o m, each pixel is
measu ed
se e al imes wi h di e en spa ial and
adiome ic esolu ion and incidence angle. p o ided good
models exis o accoun o hese a ying measu ing
condi ions, he salini y map can be ob ained by a e aging
ou all he e ie ed alues a each snap-sho . The
minimum
FOV along- ack dimension is abou 800 Km, which a
7
km/s
co esponds o 114
s,
o
146 ~nap-sho s o 0.78
s
each.
The equi alen adiome ic esolu ion
ums
ou o be hen
0.4 K, co esponding o a single-pass salini y esolu ion o
0.8
psu a 20°C o 1.6 psu a O’C. Howe e , in he cen e o
he FOV, he o al along ack dis ance
is
abou 1400
km,
gi ing 0.6 and 1.2 psu salini y esolu ions. To imp o e
hese igu es, a e aging
o
di e en pixels o sa elli e
passes is needed, which can be done due o he slow ime-
space a ia ion
o
salini y
in
open
ocean.
The ins umen will be pe iodically calib a ed using a
noise injec ion mechanism in combina ion wi h a highly
s able noise injec ion adiome e . I is expec ed ha
his
mechanism will p o ide
an
absolu e adiome ic accu acy
below
2
K.
THE
WISE
CAMPAIGN
The ange
o
incidence angles
o
MIRAS
goes
om
0’
a
nadi o app oxima ely
60°,
o pixels in he sa elli e’s
g ound- ack.
This
ansla es in o a b igh ness empe a u e
2553
ange a e ical and ho izon al pola iza ions om
50
o
150
K wi h a
small
dependence
on
wind speed (Fig. 3).
Figu e
3.
Simula ed e ical and ho izon al b igh ness
empe a u es o wind
speed
0
and
10
m/s
@
15°C &36 psu.
Fu he , he scanning con igu a ion o SMOS p esen s
new challenges:
-
Two-dimensional
imaging
o he scene, wi h a ying
incidence angles and pixel esolu ion
as
he pixel a els
ough he alias- lee ield o iew;
-
No ye well unde s ood azimu hal dependence o he
i s wo S okes pa ame e s
(T ,
Th) wi h wind
di ec ion;
-
Unknown signa u e o he
3 d
and
4*
S okes pa ame e s
(U,
V)
and
hei azimu ldele a ion dependence wi h wind
speed;
-
E ec o
sea
oam a L-band;
-
Pola iza ion
mixing
be ween e ical and ho izon al
pola iza ions due o he ela i e o ien a ion be ween he
an enna ame and he pixel's local e e ece ame;
-
Feasibili y o accu a e e ie al o
U,
and e en ually V,
assuming ha Fa aday o a ion due o he ionosphe ic
e ec s
has
been
co ec ed o by o he means.
A
Wind and Salini y Expe imen
(WISE)
is being
p epa ed o answe he i s ou challenges abo e. The
campaign will ake place a
Cusubluncu
an oil pla o m
(owned by Repsol)
40
km
away om he Eb o i e mou h.
In
his
si e high wind in ensi ies o 90
km/h
(25
m/s)
a e no
uncommon du ing Oc obe and No embe when he
campaign is o eseen o ake place
( his
yea 2000).
The equi ed measu emen s will be ob ained lom he
ollowing ins umen a ion:
-
Two L-band adiome e s:
a
ully pola ime ic
adome e T ,
U
and
V)
om he Poly echnic
Uni e si y o Ca alonia a Ba celona (Spain), and a dual
pola iza ion adiome e (Th and T ) om he Uni e si y o
Massachuse s a Amhe s (USA);
-
Th ee oceanog aphic buoys om he ICM (Spain) and
LODYC (F ance), ha will measu e
sea
su ace salini y,
sea su ace empe a u e, wind speed and wind di ec ion;
-
Po able me eo ological s a ion o a mosphmic
p essu e, empe a u e, ela i e hu midi y and
ain
a e;
-
S e eocame a lom LODYC ha will p o ide 3D
images o he sea su ace o measu e sea su ace ms slopes
-
Video images o he an enna bo esigh om a ideo
came a moun ed
on
UPC adiome e ;
-
In a ed adiome e ha will p o ide SST es ima es.
The adiome e s will be placed. in he No h side
o
he
pla o m a abou 32 m abo e he sea le el, which will also
a oid sun glin e ec s. The oceanog aphic buoys will be
moo ed in he
No h
side
so
as
o collec g ound u h da a.
Typical measu emen sequences a e:
-
Scan in ele a ion ( ixed azimu h angle), ypically
+=Oo
(No h),
s a ing
a
an
incidence angle 8i=20°,
and
ending a
an
incidence angle 3i
=
65
',
in angula s eps o
5"
o
79;
-
Scan
in azimu h a a ixed incidence angle a a
pa icula alue (Qi
=
20'-65"), s a ing a an azimu h angle
app oxima ely
I+=
120' Wes and ending a an azimu h angle
app oxima ely
4~75'
Eas in
angula
s eps o 10' o
15".
The down-welling b igh ness e:mpe a u e (a mosphe ic,
cosmic and galac ic noises) will ble de e mined &e each
measu emen by looking a 3 di e en ele a ion angles:
nadi , ho izon and zeni h.
CONCLUSIOlNS
Fo many yea s i has been possible o measu e
sea
su ace empe a u e (SST) using in a- ed sa elli e senso s.
These measu emen s ha e p o ided a g ea deal o
in o ma ion
abou he physics o he oceans. Ye
ocean
ci cula ion is dependen
on
wa e densi y
-
de e mined by
bo h empe a u e and salini y. Measu ing he sal con en o
he oceans is now becoming possible o he i s ime.
?'his
will lead o
an
imp o emen in ou knowledge o he global
ocean
ci cula ion.
REFERENCES
[l] Y. Ke ,
J.
Fon , P. Wad eu el,
A.
Camps,
J.
Ba &
I.
Co bella, F. To es, N. Dao, M. Vall.llosse a, G.
Caudal,
"Nex Gene a ion Radiome e s:
SMOS.
A Dual Pol L-baud
2D Ape u es Syn hesis Radiome e s
'I,
2000 IEEE
Ae ospace Con e ence, Ma ch, 2000 Mon ana, USA.
[2]
Camps, A., I. Co bella,
J.
Ba a F. To es, "Radiome ic
Sensi i i y Compu a ion in Ape u e Syn hesis
In e e ome ic Radiome y", IIZEE T ansac ions
on
Geoscience and Remo e Sensing, h4a ch 1998, GRS-36, No
2, pp 680-685. (Co ec ion
on
GIG-36 No.5 Sep . 1998).
2554