ocean surface and subsurface studies from space-based lidar … · 2014. 1. 27. · also collects...

Ocean surface and subsurface studies from space-‐based lidar measurements

Xiaomei Lua, Yongxiang Hub

aNASA postdoctoral program, Hampton, VA, 23666, USA; bClimate Science Branch, NASA Langley research Center, Hampton, VA, 23681, USA

AMS 94th Annual Mee:ng, Feb 2-‐6, 2014, Atlanta

Ø CALIPSO Lidar—Atmosphere measurements

The CALIPSO satellite, launched successfully on April 28, 2006, can provide two-wavelength elastic backscattered signals on a nearly global scale.


!

Outline of the talk


The primary objective of the CALIPSO mission has been studying the climate impact of clouds and aerosols in the atmosphere. However, recent studies have demonstrated that CALIPSO also collects information about ocean subsurface. The objective of this study is to,

1.  Remove the effects of lidar receiver’s transient response on the attenuated backscatter

2.  Estimate the theoretical ocean surface backscatter from the empirical relation between sea surface lidar backscatter and wind speed

3.  Estimate the Ocean sub-surface lidar backscatter

1. The effects of lidar receiver’s transient response


> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <

2

where zs is surface elevation, A stands for the amplitude of backscattered signal and ! is the root mean square width of the output pulse. A peak signal !(zp) is determined at zp that contains the surface return. Note that, zp is not necessarily the same as zs. The peak signal ratio M1 is calculated [5],

M1 =

!(zp+1)" !(zp"1)!(zp )

. (2)

where !(zp-1) and !(zp+1) are the signals at range bins immediately before and after the peak signal. The altitudes zp-1 and zp+1 are corresponding to the ‘p-1’ and ‘p+1’ range bins, respectively and ‘p’ stands for the range bin where the surface return is strongest. The peak signal ratio M1 can then be rewritten as,

M1 =

2!z"(zp )

#"(z)#z

|z=z ' ,zp$1 < z ' < zp+1. (3)

where "z=zp+1-zp=zp-zp-1 is the vertical resolution of the atmospheric profiling lidar determined by the sampling rate of digitizer, "!(z)/"(z) represents first derivative of !(z) with respect to altitude z, and z ' is altitude between zp-1 and zp+1. Substituting (1) into (3), an approximate expression of M1 is given by,

M1 = ! 4"z

# 2 "R. (4)

where "R=zp-zs is the distance between the peak signal range bin center zp and the surface zs. The surface elevation can be finally determined by,

zs = zp ! "R. (5)

The super-resolution laser altimetry technique uses the peak

surface signal !(zp) to indicate a coarse resolution of the surface zp, and also uses the peak signal ratio M1 of (2) from surface backscattered signals at three range bins to indicate the distance "R by (4). The surface elevation can be finally estimated from (5).

III. APPLICATION TO THE CALIPSO LIDAR MEASUREMENTS For the CALIPSO lidar system, the diode-pumped Nd:YAG

laser produces a pulse width of about 20 ns at a pulse repetition rate of 20.16 Hz. Beam expanders in the transmitter subsystem reduce the angular divergence of the transmitted laser beam to produce a beam diameter of 70 m at the Earth’ surface (corresponding to a nominal laser beam divergence of about 100 !rad). The receiver subsystem utilizes fourteen bits, 10 MHz analog-to-digital convertors (ADCs) equivalent to the initial 15 m vertical resolution at both 532 nm and 1064 nm channels. To conserve bandwidth, the data acquired between 0.5 km below and 8.3 km above mean sea level are subsequently averaged on board the satellite to 30 m vertical resolution at 532 nm and 60 m at 1064 nm [6]. The 130 !rad receiver field of view corresponds to a 90 m footprint on the Earth’s surface. The data source from United States Geological Survey (USGS) National Elevation Database (NED) is a high resolution digital elevation map determined from photogrammetric mapping and airborne lidar elevation data [9]. The 1 arc-second (~30 m horizontal resolution) NED data are downloaded from USGS’ Seamless Data Distribution System [9]. The 70 m diameter of the transmitted laser beam at the Earth’ surface covers more than two adjacent surface elevations (corresponding to 60 m diameter) of NED data. In order to compare with the surface elevation from CALIPSO lidar measurements, the surface elevations from NED data are averaged to the CALIPSO’s 90 m footprint.

The strongest of the CALIPSO backscatter signals are generated by ocean and land surfaces that are covered by snow or ice [5]. In the 532nm parallel channel, the peak signals from snow and ice surfaces under clear skies are so strong that they usually saturate the detectors (or digitizers). Unlike the parallel component, the perpendicular component of the ground returns for most land and ocean surfaces are generally not saturated. As a result, only the data from CALIPSO 532nm perpendicular channel were used in this study. For the satellite laser altimetry, the atmosphere forward scattering changes the shape of the surface return waveforms, and the altitude bias induced by the forward scattering for low-level cloud conditions can be of tens of cm [10]. For CALIPSO data analysis in this study, only the clearest atmospheric cases with lidar integrated attenuated backscattering coefficient of the column above the surface lower than 0.01 sr-1 [11] have been selected.

Equation. (4) is valid when surface returns and the filter’s impulse response can be assumed as Gaussian distribution. In order to obtain high resolution of surface elevation from CALIPSO lidar measurements by the new technique, accurate relationship between M1 and "R is required. Hu et al [5] estimates the relationship between M1 and "R from CALIPSO’s transient response (red line in Fig. 2). Here, the polynomial fitting is used to obtain the relationship between M1 and "R.

Ideal DetectorGain=G

Low pass filterh(t)

Received Backscattered

signal P(t)GP(t)

Analog to Digital

converter (ADC)

Range Assignment

!(ti)

!(t)=GP(t)*h(t)

Range bin zi

Attenuated backscatter !(zi)

Fig. 1. Receiver subsystem for the atmospheric profiling lidar.

This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication.The final version of record is available at http://dx.doi.org/10.1109/LGRS.2013.2256876

U.S. Government work not protected by U.S. copyright.

IJ

al

Cb 0.6

V0.2

601 _______)

40 20 0

Input pulse:P(z)Filterresponse:h(z)Transient response: p(z)

-20 -40 -60 -80Range (m)

!"#$%&'()*+,-./*/010,20/3*,)*4)0-*50/0*67/*-0879):/.:,79;*<,:5*.9*,9=4:*=4+)0*>,-:5*2./?,9@*A0:>009*BC*9)*.9-*DC*9)E*

:50* =4+)0* >,-:5* 76* 74:=4:* 6/78* :50* 6,+:0/* 15.9@0)* A0:>009* BCF* 9)* .9-* BCB* 9);* G50* >,-0* A.9-H=.))* +7>H=.))* 6,+:0/)*

),@9,6,1.9:+?*A/7.-09*:50*,9=4:*=4+)0*%IJ3*I.)*)57>9*,9*K,@;*B3;*

*

K,@4/0*B;*G50*,9=4:*=4+)0*%IJ3*IA+40*+,903E*:50*,8=4+)0*/0)=79)0*76*+7>H=.))*6,+:0/*5IJ3*I/0-*+,903*.9-*:50*-0:01:7/()*:/.9),09:*

/0)=79)0* IJ3*IA+.1L*+,903M*:50*14/20)*./0*)1.+0-*:7*:50,/*=0.L*2.+40);*

!78=./0-*>,:5*:50*.:87)=50/,1*=/76,+,9@*+,-./E*:50*+.)0/*.+:,80:/?*)?):08*790*76*:50*+0.-,9@*:0159,N40)*67/*):4-?,9@*:50*+.9-*

)4/6.10* :7=7@/.=5?* 4)4.++?* 08=+7?0-* 20/?* 6.):* 20/:,1.+* ).8=+,9@* /.:0* 76* "O!* ,9* 7/-0/* :7* .15,020* 5,@5* +.9-* )4/6.10*

/0)7+4:,79FP;* K7/* 0Q.8=+0E* :50*R07)1,0910*#.)0/*"+:,80:/?* &?):08* IR#"&3* .A7./-* :50* $10E* !+74-* .9-* +.9-*S+02.:,79*

&.:0++,:0*I$!S&.:3E*,:)*FCTU*98*/010,20/*-,@,:,J0)*:50*)4/6.10*A.1L)1.::0/0-*),@9.+)*.:*F*RVJE*>5,15*.++7>)*.9.+?),)*76*:50*

)4/6.10*01570)*>,:5*.*:,80*/0)7+4:,79*76*F*9)*.9-*17//0)=79-,9@*/.9@0*/0)7+4:,79*76*FW*18*67/*0.15*-,@,:,J0-*).8=+0FX;*

G50*/010,20/*:/.9),09:*/0)=79)0* IJ3*.9-*:50*).8=+0-*),@9.+)* IJ,3*>,:5*6.):*.9-*)+7>*).8=+,9@*)=00-*./0*)57>9*,9*K,@;*DE*

>50/0*:50*)4/6.10*0+02.:,79*,)*.))480-*.:*C*8;*K7/*:50*+.)0/*.+:,80:/?*>,:5*6.):*).8=+,9@*)=00-E*:50*+71.:,79*76*:50*=0.L*

090/@?* I7/*>.2067/8*109:/7,-3* ,)* :50* ,9-,1.:,79*76* :50* )4/6.10* I.)* )57>9* ,9* :50*8,--+0*=.90+*76*K,@;*D3;*V7>020/E*.)*

)57>9*,9*+7>0/*=.90+*76*K,@;*D*67/*:50*.:87)=50/,1*=/76,+,9@*+,-./*>,:5*/0+.:,20+?*)+7>*).8=+,9@*)=00-E*:50*+.9-*)4/6.10*

0+02.:,79*-,/01:+?*6/78*:50*+71.:,79*76*:50*=0.L*A.1L)1.::0/0-*090/@?*I7/*>.2067/8*109:/7,-3*.+790*5.20*17./)0/*20/:,1.+*

/0)7+4:,79;*G50/0*,)*.*-,):.910* Y*A0:>009*:50*)4/6.10*.9-*:50*=0.L*),@9.+*/.9@0*A,9*109:0/E*>5,15*1.9*A0*4)0-*:7*@0:*

5,@5*/0)7+4:,79*76*)4/6.10*0+02.:,79*6/78*.:87)=50/,1*=/76,+,9@*+,-./*80.)4/0809:)FP;*

O,660/09:*)0N40910)*76*-,@,:,J0-*),@9.+* IJ,3*1.9*A0*7A:.,90-*>509*"O!*).8=+0-*.9-*N4.9:,J0-*:50*.9.+7@*),@9.+* IJ3*.:*

-,660/09:* :,80;* K,@4/0* U* )57>)* :50* ).8=+0-* ),@9.+)* IJ,3*>,:5* :50* ).8=+,9@* :,8,9@* 76*"O!*-7>9>./-* I4==0/* =.90+3E*

0Q.1:+?* 17,91,-09:* I8,--+0* =.90+3* .9-* 4=>./-* I+7>0/* =.90+3* >,:5* :50* :/.9),09:* /0)=79)0;* ")* )57>9* ,9* K,@;* UE* ,6* :50*

).8=+,9@*:,8,9@*76*"O!*,)*0Q.1:+?*17,91,-09:*>,:5*:50*-0:01:7/()*:/.9),09:*/0)=79)0*I8,--+0*=.90+3E*:50*/.9@0*A,9*109:0/*

76*=0.L*),@9.+* IJ=3*.:*/.9@0*A,9*Z=[*>,++*,9-,1.:0*:50*+.9-*)4/6.10*17//01:+?;*V0/0*\=(*):.9-)*67/*:50*/.9@0*A,9*>50/0*:50*

)4/6.10*/0:4/9*,)*):/79@0):;*<509*:50*).8=+,9@*:,8,9@*,)*4=>./-*I+7>0/*=.90+3E*:50*/.9@0*A,9*109:0/*76*=0.L*),@9.+* IJ=3*

.:*/.9@0*A,9*Z=[*>,++*A0*5,@50/*:5.9*:50*0Q.1:*)4/6.10*+71.:,79E*.9-*),84+:.9074)+?*:50*8.@9,:4-0*76*).8=+0-*),@9.+)* IJ=H

F3*.:*/.9@0*A,9*Z=HF[*,880-,.:0+?*A067/0*:50*=0.L*),@9.+*>,++*A0*/0-410-*.9-*:50*8.@9,:4-0*76*).8=+0-*),@9.+* IJ=]F3*.:*

/.9@0*A,9* Z=]F[* ,880-,.:0+?* .6:0/* :50*=0.L* ),@9.+*>,++* A0* ,91/0.)0-;*<509* :50* ).8=+,9@* :,8,9@* ,)* -7>9>./-* I4==0/*

=.90+*76*K,@;*U3E*:50*17//0)=79-,9@*15.9@0)*67/*:50*).8=+0-*),@9.+)* IJ=HF3*.9-* IJ=]F3*>,++*A0*17920/)0-;*K,@4/0;*W*)57>)*

:50* /0+.:,79)5,=*.879@* :50* ).8=+0-*),@9.+)* IJ,3* .:* :5/00* /.9@0*A,9)*J=HFE* J=*.9-*J=]FE*>50/0* :50*?H.Q,)* ,)* :50* ),@9.+*.:*

/.9@0*A,9* ,880-,.:0+?* .6:0/* :50* =0.LE* IJ=]F3E* -,2,-0-*A?* :50* =0.L* ),@9.+E* IJ=3;*G50*QH.Q,)* ,)* :50* ),@9.+* .:* /.9@0*A,9*

,880-,.:0+?*A067/0*:50*=0.LE* IJ=HF3E*-,2,-0-*A?*:50*=0.L*),@9.+E* IJ=3;*

!"#$%&#'&(!)*&+#,%&--./&&--./012/

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/22/2013 Terms of Use: http://spiedl.org/terms

For a profiling lidar on a spacecraft platform (e.g., CALIOP), the receiver subsystem is illustrated in Fig.1. The lidar backscattered pulse P(z), which is a function of time t or altitude z (z=ct/2, c the speed of light), is collected by the telescope and focused onto a detector. The detector is modeled as an ideal detector with its gain of G, followed by a low-pass filter with an impulse response given by h(z).

The output of low-pass filter is sampled and quantized by an analog-to-digital converter (ADC) or a digitizer. Data is generally sampled sequentially with equal space interval corresponding to the vertical resolution of the lidar. β(zi) is the sampled signal, where zi is altitude corresponding to the ith range bin and i is a positive integer.


Because the CALIOP 532 nm laser pulse cannot easily penetrate the land surface, the lidar backscatter signal from a land surface goes quickly from a small value to a very large value and then quickly back to zero under ideal conditions, and should be distributed in single vertical range bin (30 m). However, due to the PMT’s noise tail effect and low-pass filter’s broadening effect , the strong land surface return was spread by the CALIOP instrument transient response over several adjacent range bins. Therefore, a hard land surface should be a good target for studies of the CALIOP receiver transient response function.

The CALIOP’s transient response function F can be obtained from land surface in twelve adjacent range bins as fellows:

F(z j ) =! '(z j )

! '(zi )i= p"1

i= p+10

#( j = 1,2,3,...,12)

where the twelve range bins starting from the one range bin before the peak to the tenth range bin after the peak surface return, and β'(zi) is the attenuated backscatter of each bin, i is the range bin number and p is the range bin of the peak surface return.

J. Li et al.: New method for retrieval of the extinction coefficient of water clouds 5

10−4 10−3 10−2 10−1 100

0

50

100

150

200

250

300

350

400

CALIPSO Transient Response for surface

Rang

e (m

)

P532 Channel

West coast US at July/2006West coast US at October/2006West coast SA at January/2007

10−4 10−3 10−2 10−1 100

0

50

100

150

200

250

300

350

400


Rang

e (m

)

S532 Channel


10−4 10−3 10−2 10−1 100

0

50

100

150

200

250

300

350

400


Rang

e (m

)

T532 Channel


Fig. 2. Transient response of CALIOP derived from the land surface return at different

months and different regions for three channels.

31

Fig. 2. Transient response of CALIOP derived from the land surface return at different months and different regions for three channels.

... = ... (6)n�

i=1βicorrected×Fn−i+1= βi−1

current (n = 1,2,3,4,...) (7)

After obtaining the transient response function F ofCALIOP, we can use it in conjunction with the currentlidar signal to retrieve the corrected water cloud attenu-ated backscatter signal by reversing the convolution pro-cess described by Eqs. (4)–(7), which corresponds to a de-convolution process. Before the de-convolution process, wemust do some horizontal averaging of the vertical lidar pro-files (using for example, 30 profiles) in order to eliminatepossible negative values in the water cloud profiles due tofilter noise. Then, we may start the de-convolution pro-cess from several bins (here, we only use one bin) whichhave very weak air backscatter value above the water clouds.For example, in Eq. (4), β0current consists of weak backscat-ter from the air just above the cloud, as well as backscat-ter from the first bin within the water cloud. Comparedto the backscatter from the first cloud bin, the backscatterfrom the air just above the cloud is very weak and can beneglected. Thus, β0current is the backscatter signal from thefirst bin within the water cloud, and β1corrected is the correctedbackscatter value of first bin of the water cloud profile, andβ1current is the current backscatter value of first bin of the wa-ter cloud profile. By continuing this de-convolution process,eventually, the corrected backscatter signals of all bins canbe derived.Figure 3 shows the cloud attenuated backscatter signal re-

trieved beneath the water cloud peak return and the observedattenuated backscatter signal by CALIOP. The red line isobserved (current) water cloud attenuated backscatter signaland the blue line is the retrieval (corrected or real) cloud sig-nal. The results show that the transient response of CALIOPPMTs can affect the vertical distribution (that is, the wave-form) and magnitude of the water cloud attenuated backscat-ter signal. After the de-convolution process, the slope of the

exponential decay of the water cloud attenuated backscatter,may be obtained by using a simple linear fit to the severalrange bins underneath the peak of the water cloud lidar re-turn and the peak return bin itself. According to Eq. (2), theextinction coefficient of the low-level water cloud top thuscan be derived from the slope and multiple scattering factorof the water cloud.

3 Results

3.1 Comparison of extinction coefficients derived fromdifferent methods

Hu et al. (2007a) derived the mean extinction coefficient σ

of water cloud top by combining the cloud effective radiusRe reported by MODIS with the lidar depolarization ratiosmeasured by CALIPSO:

σ = (Re

Re0)1/3

�1+135 δ2

(1−δ)2

�(8)

where Re0 equals 1 µm, and δ is the layer-integrated depolar-ization ratio from CALIPSO Level 2 cloud products. Equa-tion (8) is derived from Monte Carlo simulations that incor-porate the CALIPSO instrument specifications, viewing ge-ometry, and footprint size. This method (hereafter, we callit “Hu’s method”) needs collocated water cloud droplet sizesretrieved from MODIS 3.7-µm data for CERES (Minnis etal., 2006). The number of photons scattered into the forwarddirection increases with particle size. Thus, the chance ofa photon at the near-infrared 3.7-µm wavelength being ab-sorbed rather than backscattered to space increases with size.For the same optical depths, water clouds with larger dropletsare darker in the near-infrared wavelengths. The effectivedroplet radius derived from the absorption at 3.7-µm reflectsthe average size information from the very top part of wa-ter clouds (Platnick, 2000), with a vertical penetration depthsimilar to the CALIPSO lidar signal. So, Hu’s method is a

www.atmos-chem-phys.net/11/1/2011/ Atmos. Chem. Phys., 11, 1–15, 2011

!


Actually, the current observed attenuated backscatter signal β'm(z) is a result from a convolution between the correct attenuated backscatter β'c(z) and the CALIOP transient response function F. This convolution process can be described mathematically as follows,

F(z2 ), F(z1),0,!!!!!!,0F(z3), F(z2 ), F(z1),!!!!,0!!!!!!!!!!!!!F(zn+1), F(zn ), F(zn!1),!, F(z2 )

"

#

$$$$$

%

&

'''''

(

) 'c (z1)

) 'c (z2 )

") 'c (zn )

"

#

$$$$$

%

&

'''''

=

) 'm(z1)

) 'm(z2 )

") 'm(zn )

"

#

$$$$$

%

&

'''''

With the values of the transient response functions, we can retrieve the correct attenuated backscatter signal by de-convolution process

! 'c = M "1 #! 'm

M!c' = !m

'That is,

Ocean subsurface observed and corrected depolarization ratio


Fig. 3. CALIOP Ocean subsurface observed (upper panel) and corrected depolarization !w (lower panel).

Fig. 3. CALIOP Ocean subsurface observed (upper panel) and corrected depolarization !w (lower panel).

The integrated ocean subsurface depolarization ratios δw, defined as the ratio of the cross-polarized signal to co-polarized signal integrated from the first range bin to the tenth range bin below the surface

Global distribution (2o by 2o) of CALIOP Ocean subsurface observed and corrected depolarization ratio δw. Color code is the value of δw.

! w = " '# (zi )

i=1

10

$ / " '!(zi )i=1

10

$

Ocean observed and corrected total depolarization ratio


Fig. 4. The column-integrated depolarization ratio !T that includes surface and subsurface backscatter

3. Ocean Subsurface results Detection of subsurface returns from CALIOP was demonstrated. Evidence of subsurface scattering was observed as an increased depolarization with depth. These features were all correlated with near-surface chlorophyll concentrations. An increase in the depolarization was also seen at a depth of 50 m under certain conditions, suggesting that chlorophyll concentration at that depth could be estimated if an appropriate retrieval technique can be developed.

The empirical relation between sea surface lidar backscatter and wind speed [12] was used to estimate

the theoretical ocean surface backscatter. Then the two-way atmospheric transmittance was obtained as the ratio between the CALIPSO measured ocean surface backscatter and the theoretical one. The ocean subsurface lidar backscatter profile is finally derived from the subsurface attenuated backscatter divided by

Fig. 4. The column-integrated depolarization ratio !T that includes surface and subsurface backscatter

3. Ocean Subsurface results Detection of subsurface returns from CALIOP was demonstrated. Evidence of subsurface scattering was observed as an increased depolarization with depth. These features were all correlated with near-surface chlorophyll concentrations. An increase in the depolarization was also seen at a depth of 50 m under certain conditions, suggesting that chlorophyll concentration at that depth could be estimated if an appropriate retrieval technique can be developed.

The empirical relation between sea surface lidar backscatter and wind speed [12] was used to estimate

the theoretical ocean surface backscatter. Then the two-way atmospheric transmittance was obtained as the ratio between the CALIPSO measured ocean surface backscatter and the theoretical one. The ocean subsurface lidar backscatter profile is finally derived from the subsurface attenuated backscatter divided by

The integrated total depolarization ratio δT that includes both surface and subsurface backscatter

Global distribution (2o by 2o) of CALIOP Ocean observed and corrected total depolarization ratio δT. Color code is the value of δT.

1. The effects of lidar receiver’s transient response


•  Affect the vertical distribution (that is, the waveform) and magnitude of the lidar backscatter signal and the depolarization ratio.

•  Affect the values of the integrated subsurface depolarization δw and total depolarization ratio δT

As a result, the CALIOP transient response can

Thus, the effects of CALIOP receiver transient response on the attenuated backscatter should be removed first.

2. Estimate the theoretical ocean surface backscatter


The theoretical ocean surface backscatter can be estimated as,

!532

s = 0.02094"# 2 cos4$

exp[% tan2$2# 2 ] Where θ is CALIOP’s off nadir pointing angel (0.3 or 3

degree) and σ2 is the wave slope variance, which is a function of AMSR-E wind speed [1].

Reference 1.  Y. Hu, K. Stamnes, M. Vaughan, J. Pelon, C. Weimer, D. Wu, M. Cisewski, W. Sun, P. Yang, B. Lin, A. Omar, D. Flittner, C.

Hostetler, C. Trepte, D. Winker, G. Gibson, and M. Santa-Maria, "Sea surface wind speed estimation from space-based lidar measurements," Atmos. Chem. Phys., vol. 8, pp. 3593-3601, 2008

Then, the transmittance of the overlying atmosphere can be estimated as the ratio between the CALIOP ocean surface backscatter and the theoretical one,

T532

2 =!532

s '

!532s

When the effects of CALIOP detector’s transient response on the attenuated backscatter profile were removed, the CALIOP correct attenuated backscatter signal from ocean surface and subsurface is

!532' = !532T532

2where T is the one-way atmospheric transmittance along the lidar look direction, β is the volume backscatter coefficient.

Finally, the ocean subsurface backscatter can be obtained as,

!532

u =!532

u '

T5322

3. Estimate the Ocean sub-surface lidar backscatter


532nm Perpendicular: Subsurface Par:culate BackscaFer from 532nm Cross Polariza:on Signal

• For a linearly polarized incident lidar beam (e.g., CALIOP), spherical par:cles, Rayleigh scaFering, and reflec:on at the ocean surface do not contribute significantly to cross polariza:on

• Cross polariza:on (measured by the perpendicular channel) is dominated by backscaFering of non-‐spherical par3cles

e.g., cloud ice crystals in atmosphere plankton and other non-‐spherical

par3cles in the water

3. Ocean subsurface results


Global distributions (2o by 2o) of MODIS chlorophyll a concentration, C (mg/m3) and Particulate Organic Carbon, POC (mg/m3). Color code is the decimal logarithm of C and POC.

Global distribution (2o by 2o) of CALIOP Ocean subsurface backscatter

Relationship between the obtained CALIPSO subsurface backscatter γ and MODIS chlorophyll-a C and Particulate Organic Carbon, POC


log10 ! = 0.17 log10 C " 3.7

! = 1.35#10"4 log10 POC "8.1#10"5

We found an interesting relation between integrated ocean subsurface backscatter γ(/sr) and chlorophyll-a concentration C (mg/m3), and relation between γ(/sr) and POC (mg/m3) as,

The relations indicate a potential of CALIOP lidar for quantifying global chlorophyll-a and POC concentrations, which will open up a new application for the space-based CALIOP lidar

Summary


Besides the primary cloud and aerosol data products such as cloud types, cloud physical properties, aerosol type and aerosol optical properties, CALIOP profiling lidar measurements also provides a lot more information for ocean studies, e.g,

•  Ocean sub-surface particulate backscatter •  Ocean sub-surface depolarization ratio •  Estimate chlorophyll-a and POC concentrations

Contact Email: [email protected] [email protected]

Any ques:on? Thank you for your aFen:on


ocean surface and subsurface studies from space-based lidar … · 2014. 1. 27. · also collects...

Documents