Investigation of ice particle habits to be used for ice cloud remote sensing for the GCOM-C satellite mission

Investigation of ice particle habits to be used for ice cloud remote sensing for the GCOM-C satellite mission H. Letu, H. Ishimoto, J. Riedi, T. Y. Nakajima, L. C.-Labonnote, A. J. Baran, T. M. Nagao, and M. Skiguchi Research and Information Center (TRIC), Tokai University, 4-1-1 Kitakaname Hiratsuka, Kanagawa 259-1292, Japan Meteorological Research Institute, Nagamine 1-1, Tsukuba 305-0052, Japan Laboratoire d’Optique Atmosphérique, UMR CNRS 8518, Université de Lille 1-Sciences et Technologies, Villeneuve d’Ascq, France Met Office, Fitzroy Road, Exeter, EX1 3PB, UK Earth Observation Research Center (EORC), Japan Aerospace Exploration Agency (JAXA), 2-1-1 Sengen Tsukuba-shi, Ibaraki 305-8505, Japan Tokyo University of Marine Science and Technology, Tokyo 135-8533, Japan

confirmed that the SAD of small bullet rosette and all sizes of voronoi particles has a low angular dependence, indicating that the combination of the bullet-rosette and Voronoi models are sufficient for retrieval of the ice cloud spherical albedo and optical thickness as an effective habit models of the SGLI sensor.Finally, SAD analysis based on the Voronoi habit model with moderate particles ( r = 30 µm) is compared to the conventional General Habit Mixture (GHM), Inhomogeneous Hexagonal Monocrystal (IHM), 5-plate aggregate and ensemble ice particle model.It is confirmed that the Voronoi habit model has an effect similar to the counterparts of some conventional models on the retrieval of ice cloud properties from space-borne radiometric observations.

Introduction
Ice clouds play an important role in the radiation balance of the Earth's atmospheric system through interaction with solar radiation and infrared emission (Liou, 1986;Baran, 2012).However, there are still large uncertainties in characterizing their microphysical and optical properties.This is because they consist of ice particles with a wide range of habits and sizes (C.-Labonnote et al., 2000;Forster et al., 2007;Baran et al., 2009;Cole et al., 2014;Yang et al., 2015).Different ice particle habits have varying single-scattering characteristics, resulting in different radiative properties.The only feasible way of inferring ice cloud properties on a global scale is to use satellite observations.In practice, an ice model is chosen, which may consist of a single habit or a mixture of habits, and look-up tables (LUTs) for ice cloud reflection and transmission characteristics are computed for a range of input optical properties, particularly, optical thickness, cloud temperature, and effective particle size.The LUTs and a fast radiative transfer model are used subsequently for global operational retrievals.Thus, the choice of an ice model for a given satellite mission deserves rigorous investigation.
In this paper, the present study is to better understand the performance of several ice cloud habit models, in conjunction with applications to the Global Change Observation Mission-Climate (GCOM-C) satellite mission.Figures
Based on one or more of these scattering models, single-scattering property libraries have been developed for certain habits.Hess et al. (1994) developed the singlescattering properties database for hexagonal plates and columns with a random orientation at wavelengths between 0.35 and 3.7 µm.Yang et al. (2000) developed a scattering and absorption property database for various ice particle habits with random orientation using the FDTD and IGOM methods, at wavelengths between 0.2 and Figures

Back Close
Full ing the FDTD, T-matrix, IGOM, and Lorenz-Mie theory.Recently, Yang et al. (2013) released a full set of scattering, absorption, and polarization properties assuming random orientation for a set of 11 habits at a number of wavelengths, ranging between 0.2 and 100 µm.The habits include droxtals, prolate spheroids, oblate spheroids, solid and hollow columns, and compact ice aggregates which were composed of eight solid hexagonal ice columns (hereafter, "8-column aggregate", hexagonal plates, small spatial aggregates composed of 5 plates (hereafter, "5-plate aggregate"), large spatial aggregates, composed of 10 plates, and solid and hollow bullet-rosettes.This library is based on the Amsterdam discrete dipole approximation (ADDA), T-matrix, and IGOM methods.Based on this updated library, Baum et al. (2014) developed a new set of bulk scattering and absorption models with habit mixtures for use in ice cloud radiative transfer calculations and retrieval of the ice cloud properties from remote sensing measurements.The ice crystal single-scattering databases of Baran and Francis (2004) and Baran et al. (2014) have been constructed from numerical simulations of individual ice crystals by using a combination of methods described in those papers.
This study is aimed at identifying an optimal choice of ice habits for the Global Change Observation Mission (GCOM-C) satellite mission.It is useful to provide a brief summary of what other teams have chosen for their operational products.Numerous articles have investigated the use of effective ice particle habits derived from various ice habit models and remote sensing measurements from multi-angles for use in cloud parameter retrievals (Baran et al., 1998(Baran et al., , 1999(Baran et al., , 2003;;Chepfer, 1998;C.-Labonnote et al., 2000;Chepfer et al., 2001;Masuda et al., 2002;Knap et al., 2005;Sun et al., 2006;Baran and Labonnote, 2006).Baran et al. (2007) sensitive to the ice particle habit and roughness, at least for ice clouds having an optical thickness larger than 5. Chepfer et al. (2002) investigated the effective ice particle habits using multi-angle and multi-satellite methods from visible reflectance satellite measurements.There is increasing evidence that the ice particle model should contain some degree of surface roughness (Foot, 1998;Baran et al., 2001Baran et al., , 2003;;Ottaviani et al., 2012a;van Diedenhoven et al., 2012;Cole et al., 2013Cole et al., , 2014)).Yi et al. (2013) reported that the general habit mixture model (GHM, Baum et al., 2011Baum et al., , 2014) provided significant differences in the shortwave cloud radiative effect in the National Center for Atmosphere Research Community Atmosphere Model (NCAR CAM 5).Baran and Labonnote (2007) developed an ensemble ice particle model for cirrus using various habits including single hexagonal columns, a six-branched bullet-rosette, and more complex three-branched, five-branched, eight-branched, and ten-branched particles.Baran et al. (2014) demonstrated that it is possible to simulate measured cirrus radiances from the UV to microwave frequencies by using the same microphysically consistent habit mixture model throughout the spectrum.At solar wavelengths, both Baran and Labonnote (2007) and Baran et al. (2014) showed that featureless phase functions best fitted their multi-angle solar radiance measurements.Moreover, Baran et al. (2015) investigated the relationship between the habit model of the cirrus cloud particles and the atmospheric relative humidity.These studies suggest that whatever ice model is employed, it should have a featureless phase function at solar wavelengths.Ishimoto et al. (2012b) developed a new habit of complex and highly irregular shapes called Voronoi aggregate, which was based on the ice particle images of convective ice clouds from in-situ measurements.The phase function of the Voronoi habit varies smoothly with scattering angle, which is similar behavior found from assuming severe surface roughness or including bubbles within the particle or a combination of included bubbles and surface roughness.However, the use of the Voronoi habit model for retrieval of the ice cloud optical thickness has not been investigated yet.
In this study, the single-scattering properties of the various ice particle models were calculated for developing the ice cloud property products of the GCOM-C Second Gen-Figures

Back Close
Full eration Global Imager (SGLI) satellite sensor.FDTD (Ishimoto et al., 2012a), GOIE (Ishimoto et al., 2012a) and GOM (Masuda et al., 2012) methods are used to calculate the light scattering properties for five ice habits including columns, plates, droxtals, bullet-rosettes, and Voronoi, for a set of SGLI wavelengths.The SAD analysis is performed to investigate the optimal habit(s) using the POLDER-3 data.Furthermore, the results of the SAD analysis of the Voronoi habit model are compared with the conventional GHM, the IHM model and 5-plate aggregate model of Yang et al. (2013).

Single-scattering properties of the ice particle for the SGLI sensor
Single-scattering properties for the five ice particle habits are calculated for the SGLI observation channels.The single-scattering properties are used to determine the optimal ice particle habits using the SAD method.The SGLI is the successor sensor to the Global Imager (GLI) on board ADEOS-II, which takes measurements at wavelengths ranging from the near ultraviolet to the thermal infrared (IR).The first satellite GCOM-C1 is scheduled for launch in 2017 by the Japan Aerospace Exploration Agency (JAXA).The GCOM-C mission intends to establish a long-term satellite-observing system to measure essential geophysical parameters on the Earth's surface and in the atmosphere on the global scale, to facilitate the understanding of the global radiation budget, carbon cycle mechanism and climate change (Imaoka et al., 2010).As shown in Table 1, SGLI has 19 channels, including two polarization channels at visible/nearinfrared wavelengths.A detailed description of the SGLI is reported by Imaoka et al. (2010), Nakajima et al. (2011), andLetu et al. (2012).Four of the ice particle habits (hexagonal columns, plates, bullet-rosettes, and droxtals) employed in this study were chosen by referring to MODIS collections-5 ice particle model (Baum et al., 2005) and ice cloud in-situ measurement data.The habits shown in Fig. 1 are defined with the same parameters (semi-width, length, aspect ratio and maximum dimension) as employed in the scattering properties database by Yang et al. (2000Yang et al. ( , 2005)) of Wigner-Seitz cells from a 3-D mosaic image of the ice cloud microphysical data (Ishimoto et al., 2012b); this habit is different from the aggregate model used in the scattering database reported by Yang et al. (1998bYang et al. ( , 2013)).Spatial Poisson-Voronoi tessellations were used to determine the complex structure of the ice particles for the 3-D mosaic image.The geometry of each cell in the Voronoi tessellation is defined and is based on the method by Ohser et al. (2000).
A combination of the FDTD, GOIE, and GOM methods is employed to calculate the single-scattering properties of these five ice habits for a wide range of size parameters (SZP).The refractive index of ice published by Warren and Brandt (2008) is used in the computations.As shown in Table 2, the FDTD method is used to calculate the single-scattering properties of ice particles with small size parameters (SZP < 50).The GOIE and GOM methods are employed for calculating the scattering properties of the ice particle with medium and large size parameters, respectively.The wavelength selected for detailed calculations is determined by optimizing the results of the scattering database for the SGLI channels (Letu et al., 2010).Calculations are performed at 27 spectral wavelengths (λ) from the visible to the IR spectral region in the SGLI channels shown in Table 2.The effective particle radius ( r) ranges from 0.7 to 533 µm and the SZP ranges from 0.35 to 6098.The r is defined as a single particle radius of equivalent volume sphere.The SZP is given as, Consideration of the edge effect (Bi et al., 2010, Bi andYang, 2014) is important for calculating the extinction efficiency (Q ext ) and absorption efficiency (Q abs ) by the GOIE method when the size parameter is less than 1000.The treatment of the edge effect is based on the method proposed by Ishimoto et al. (2012a).Correction coefficients are calculated in this study from comparison results of the FDTD and GOIE as shown in Eq. ( 2): Full where Q ext/GOIE and Q abs/GOIE are the extinction efficiencies, respectively, calculated by the GOIE method.K 1 and K 2 are the coefficients of the edge-effect contribution.These coefficients are applied to correct the Q ext and Q abs of large particles calculated using GOIE, which is calculated by comparing Q ext and Q abs obtained from FDTD and GOIE for maximum extension from the center of mass ranging from 30 to 60 µm.Introduction

Conclusions References
Tables Figures

Back Close
Full  (2000) and Baran et al. (2006) proposed the SAD method for testing the phase function of the various ice particle models using POLDER observational data with multi-viewing angles.For investigating the phase functions (P 11 ) of the different ice crystal models for retrieving the cloud microphysical properties, the cloud spherical albedo as a function of scattering angle is required.For calculating the cloud spherical albedo, bi-directional reflection is first determined by: R cld (τ, r e , ω, where τ, r e and ω are the cloud optical thickness, effective cloud particle radius and bulk-scattering albedo, respectively; u, and u 0 are cosines of the satellite and solar zenith angles, respectively; φ is the relative azimuth angle between satellite and the sun; L obs is the reflected solar radiance observed by satellite; F 0 is the solar flux density.Based on Eq. ( 3), τ and r e can be retrieved from L obs using a look-up table (LUT) method (Nakajima and King, 1990;Nakajima and Nakajima, 1995) where n r is the number size distribution as a function of r.In Eq. ( 4), the values of R cld over the range of viewing geometries is required for calculation of the cloud albedo.
From that, cloud optical thickness is retrieved from L obs data using the LUTs calculated from scattering property of various ice particle habits.Then, L obs data over the range of satellite viewing geometries is simulated using a radiative transfer model.Based on these calculations, R cld is calculated using Eq. ( 3).Number of the cloud spherical albedo (S(θ)) can be calculated from the L obs in each pixel of the POLDER measurements with various scattering angles (θ) to investigate the P 11 element of the ice particle models.Total observation number (N) of the L obs , with various θ, is up to 16, which is limited to the viewing geometries of the measurements.Baran et al. (2006) assumed that if the scattering phase function of the ice particle model is correct, retrievals of the τ(θ) and S(θ) in each direction should be the same and the SAD, as shown in Eq. ( 8), should be 0. S, SAD and θ are given as, where u 0 and u are the solar and satellite zenith angles, respectively.The steps to apply the SAD analysis to POLDER-3 measurements are as follows: 1. Calculate spherical albedo from the POLDER-3 measurements with 16 viewing geometries for each of the ice particle models.31675 Introduction

Conclusions References
Tables Figures

Back Close
Full 2. Perform the SAD analysis by taking the difference between the directional and the direction-averaged cloud spherical albedo.
3. Assume that the phase function for each ice particle model adequately represents the phase function for all the ice particles in each pixel of the satellite measurement, and that the retrievals of the optical thickness and spherical albedo from the POLDER measurements with different viewing geometries are the same.
When SAD is 0, the mean spherical albedo and the spherical albedo from the specific angle of POLDER-3 measurements are the same.Therefore, the criteria for selecting the optimal particle habit of the ice cloud are defined as a SAD near 0 in the 16 viewing geometries of POLDER-3, and a small angular dependence.

Characteristics of the single-scattering properties
To confirm the accuracy of the calculated single-scattering properties, the phase functions computed in this study are compared with other results.Figure 3 shows comparisons of the phase function (P 11 ) of hexagonal and spheroid particles calculated from the FDTD method with those derived from the ADDA (Bi et al., 2011) and T-matrix methods, respectively.Our FDTD results are the same as those calculated with the other methods.In addition, Ishimoto et al. (2012b) and Masuda et al. (2012) verify that the phase functions of ice particles with medium and large size parameters are the same by comparing the GOIE and GOM results, respectively.
Phase functions of the column, droxtal, plate, bullet-rosette, and Voronoi habit with various size parameters at wavelengths of 0.686 µm are given in Fig. 4. The phase functions depend on the particle habit and size parameters.There is a halo peak for the column, plate and bullet-rosette habit when the size parameter is larger than 100, as particle roughening is not applied in these calculations.For the droxtal, variation of Introduction

Conclusions References
Tables Figures

Back Close
Full the phase function is evident when the size parameter is sufficiently small (SZP < 10).
The POLDER-3 measures intensity from 16 viewing directions at scattering angles between 60-180 • ; for these scattering angles, the phase function curves of the various particles are different.The phase function of the Voronoi habit is very smooth, with features similar to those for severely roughened ice particle models and the IHM model except for the halo peak region as reported by Yang et al. (2013) and Doutriaux et al. (2000).
The asymmetry factor, single-scattering albedo, and extinction efficiency are some of the key parameters of the single-scattering properties of ice particles.Figure 5 shows the single-scattering properties of various ice particle habits at the wavelengths of 1.05 and 2.2 µm for various size parameters.The extinction efficiency increases with the size parameter up to approximately 10 and converges gradually to 2 when the size parameter is larger than 100.The maximum values of the extinction efficiency appear when the size parameter is around 10.However, the location of the maximum extinction efficiency varies with particle habit.There is a smooth peak in the asymmetry factor for the size parameter from 1 to 10.The peak of the asymmetry factor at a wavelength of 2.1 µm is larger than that at 1.05 µm.
Figure 6 shows the comparison of the satellite-observed radiance from the column ice particle model with the other four models at wavelengths of 1.05 and 2.21 µm as a function of optical thickness τ and effective cloud particle radius r e .The radiance is calculated from the RSTAR radiative transfer model (Nakajima and Tanaka, 1986;Sekiguchi and Nakajima, 2008), which is a general package for simulating radiation fields in the atmosphere-land-ocean system at wavelengths between 0.17 and 1000 µm.The solar zenith, viewing zenith, and relative azimuth angles are 40, 30, and 90 • , respectively.The satellite-observed radiance at a wavelength of 2.21 µm for the column habit is similar to the plate and droxtal models for the same r e but different at the 1.05 µm wavelength.The radiances from the column model are significantly different from those obtained with the bullet-rosette and Voronoi models, which results in different values of the inferred τ and r e .Thus, the determination of the effective ice par-Introduction

Conclusions References
Tables Figures

Back Close
Full ticle model is important for reducing the retrieval error caused by using different particle models.

SAD analysis
Figure 7 shows the SAD analysis as a function of the scattering angle, effective particle radius, and ice particle model.The SAD of the droxtal, column and plate show substantial variations in both the scattering angle and effective particle radius.The variation of SAD for the bullet-rosette model is more smoothly distributed close to 0 value of the SAD (hereafter, "zero line") than with the droxtal, plate, and column models for small (6 < r < 10 µm), medium (28 < r < 38 µm), and large particles (70 < r < 100 µm).However, the SAD peak of the bullet-rosette model varies in the scattering angle range of 140 to 160 • with medium and large particles.The SAD of the Voronoi model is closest to the zero line over the entire scattering angle range for small, medium and large particles.
Both the Voronoi and bullet-rosette model with small particles are smoothly distributed along the zero line.
Figure 8 shows the slope of the regression function (SRF) and total relative albedo difference (TRAD) of the SAD for the same five ice particle models with small, medium and large particles as shown in Fig. 7.Both values of the SRF and TRAD for small particles of the bullet-rosette, for medium and large particles of the bullet-rosette and Voronoi models, are smallest of all the single particle models considered.However, there is a changing peak of the SAD in the scattering angle range of 140 to 160 • for bullet-rosette model with medium and large particles.The SRF for large particles and TRAD for all size of particles with droxtal model are largest in the all ice particle models.As we have described in Sect.3, the effective particle habit is defined as a smallest value of the SRF and TRAD.Thus, it was confirmed that the bullet-rosette model with small particles and Voronoi model with medium and large particles are sufficiently accurate for the retrieval of the ice cloud spherical albedo and optical thickness.Thus, these models are sufficient to represent cirrus in terms of effective habit models for the purposes of the SGLI sensor.Introduction

Conclusions References
Tables Figures

Back Close
Full Ice crystals in ice cloud are complex.To simulate this complexity, we assume different values of distortion (as defined by Macke et al., 1996) and apply these to the ensemble model.Numerous previous studies have shown that the degree of distortion is an important property to consider when retrieving ice cloud optical properties from multiple-view instruments.To investigate the influence of the distortion of the ice particle model on retrieval of the ice cloud property, we performed the SAD analysis using the ensemble ice particle models with r = 30 µm, assuming a number of distortion values (see Fig. 9).The variation of SAD for the no distortion model in Fig. 9a is largest relative to the other distortion values.As a function of distortion value, there are significant variations in the SAD analysis in the scattering angle range of 60 to 80 • and 140 to 160 • .There is no obvious difference of the SAD between Fig. 9b, Fig. 9c and Fig. 9d for various degrees of the distortions.The SAD of the ice particle models with a distortion values of 0.4 with spherical air bubbles in Fig. 9e is closest to the zero line.It is implied that the models with distortion or surface roughness are best for the retrieval of the ice cloud optical property than with no distortion applied to the model.
The model with spherical air bubbles and distortion is most efficient than the models with distortion only.
Several conventional studies demonstrated that ice particle models such as ensemble ice particle model, IHM and GHM and some aggregated complex models, with rough surface are useful for the operational satellite data processing (C.-Labonnote et al., 2000(C.-Labonnote et al., , 2001;;Doutriaux et al., 2000;Baum et al., 2011Baum et al., , 2014;;Baran and Labnnote,. 2006, 2007and Cole et al., 2013).For evaluating the accuracy of the Voronoi model, we further compared the SAD of the Voronoi model with the conventional IHM, GHM, 5plate aggregate and ensemble ice particle models with r = 30 µm.As shown in Fig. 10, all of the selected models do not have any strong angular dependence.However, all the models in Fig. 10 have a rough surface except for the IHM containing spherical air bubbles and Voronoi habit.It is implied that the Voronoi habit model has a similar effect as some aggregated and mixed-habit ice particle models with roughened surfaces Introduction

Conclusions References
Tables Figures

Back Close
Full and the IHM single particle model containing air bubbles on retrieval of the ice cloud properties using remote sensing instruments.
Figure 11 shows the slope of the regression function (top panel) and total relative albedo difference (bottom panel) for the selected models in Fig. 10.The SRF for ensemble ice particle model, GHM and Voronoi model was significantly smaller than the other two models.The TRAD for all of the habit models is not significantly different from each other.However, Voronoi model is little bit smaller than the other models except for ensemble ice particle model.Voronoi model has both of the small values of SRF and TRAD, which is similar to ensemble ice particle model, indicating that SAD of the Voronoi model has a low angular dependence.

Conclusions
Ice particle single-scattering properties are investigated for potential use in the GCOM-C satellite program.The single-scattering properties of five different ice particle models (plates, columns, droxtals, bullet-rosettes, and Voronoi) are developed using the FDTD, GOIE and GOM methods.The accuracy of the single-scattering property is investigated by comparing the phase function from the FDTD method used in this study to conventional results from ADDA and T-matrix method.The FDTD phase functions are also compared with computational results from GOIE.Results indicate that the FDTDbased phase functions are consistent with results from the ADDA, T-matrix and GOIE methods and suggest that the single-scattering property database developed in this study is reliable for use in radiative transfer simulations and applications in the remote sensing of ice cloud.
The characteristics of the single-scattering property database are investigated by analyzing the asymmetry factor, single-scattering albedo, and extinction efficiency.Each of the five habits are employed to compare top-of-atmosphere radiances as a function of the optical thickness and effective particle radius.Results indicate that the satelliteobserved radiances from the column model are significantly different from those ob-Introduction

Conclusions References
Tables Figures

Back Close
Full tained with the bullet-rosette and Voronoi models when the effective particle radius is larger than 8.These results imply that the choice of ice particle habit is important for reducing the retrieval error.Furthermore, the SAD analysis is performed to determine the optimal ice particle habit for retrieving the optical thickness and cloud spherical albedo using POLDER-3 multi-angle measurements.Retrievals are performed using 589 246 pixels of the POLDER-3 observation data with a global scale over ocean from days 20-22 March, June, September, and December 2003.The following conclusions are drawn from these results.
The SADs of the droxtal and column habits show significant variations with scattering angle and effective particle radius.The variation of the SAD for small particles with bullet-rosette model is more smoothly distributed along the zero line than other habit models.
The SAD of the Voronoi model is closest to the zero line with scattering angle for all size of particles.
The bullet-rosette habit for small particles and Voronoi habit for all size of particles are most suitable for retrieving the ice cloud spherical albedo and optical thickness.
The results of the SAD analysis from the Voronoi model are compared with the result from the conventional IHM, GHM, 5-plate aggregate and ensemble ice particle model with moderate ice particle size for evaluating the efficiency of the Voronoi model on retrieving ice clouds optical properties.It is concluded that the Voronoi habit model is similar to the conventional efficient models for retrieval of the ice cloud properties using remote sensing instruments.The results of this study are not only useful for developing the ice cloud products of the GCOM-C/SGLI satellite mission, but also useful for determining the optimal ice particle habit for ice cloud remote sensing.Figures

Back Close
Full  Full  Full  Full  Full Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | . C.-Labonnote et al. (2000, 2001) and Doutriaux et al. (2000) developed models of randomly oriented hexagonal ice particles containing spherical air bubbles (inhomogeneous hexagonal monocrystal (IHM) model) for use in the ice cloud retrievals of the POLDER (POLarization and Directionality of the Earth's Reflectances) measurements.The spherical albedo difference (SAD) analysis is employed to investigate the capability of the IHM model for retrieving the optical property of ice cloud.It is illustrated that the POLDER multi-angle measurements is Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | . The Voronoi habit is determined numerically by extraction Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |is increased in the scattering angle range of 60 to 160 • and is decreased in the scattering angle range of 160 to 180 • .There is a changing peak of the number of sample pixels in the scattering angle range of 140 to 160• .Figure2bindicates the variation of the number of pixels with the different latitude.The number of pixels is changing significantly as a function of latitude and is lowest when the latitude is around 90 • N and 90 • S. C.-Labonnote et al.

Table 1 .
Specification of the SGLI.

Table 2 .
Size parameter with various particle size and calculating wavelength on the SGLI channels (FDTD, GOIE, GOM).