A41B-0425
Atmospheric and surface/cloud parameters retrieved from satellite hyperspectral infrared sounder measurements
An advanced retrieval algorithm with a fast radiative transfer model, including cloud effects, is used for atmospheric profile and cloud parameter retrieval. This physical inversion scheme has been developed, dealing with cloudy as well as cloud-free radiance observed with ultraspectral infrared sounders, to simultaneously retrieve surface, atmospheric thermodynamic, and cloud microphysical parameters. A fast radiative transfer model, which applies to the clouded atmosphere, is used for atmospheric profile and cloud parameter retrieval. A one-dimensional (1-d) variational multi-variable inversion solution is used to improve an iterative background state defined by an eigenvector-regression-retrieval. The solution is iterated in order to account for non-linearity in the 1-d variational solution. This retrieval algorithm is applied to the MetOp satellite Infrared Atmospheric Sounding Interferometer (IASI) launched on October 19, 2006. IASI possesses an ultra-spectral resolution of 0.25 cm-1 and a spectral coverage from 645 to 2760 cm-1. Preliminary retrievals of atmospheric soundings, surface properties, and cloud optical/microphysical properties with the IASI measurements are obtained and presented.
A41B-0426
Comparison of Cloud Amount between CALIPSO and General Circulation Models
General Circulation Models (GCMs) are the primary tools for climate change prediction. The greatest uncertainties in GCM simulations and future projections of climate probably arise from clouds and their interactions with radiation. CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) was launched into a sun-synchronous orbit on April 28, 2006, where it joined the A-Train constellation of four other Earth-orbiting satellites: Aqua, Aura, CloudSat and Parasol. Fielding the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) and combined with the Cloud Profiling Radar aboard CloudSat, CALIPSO is able to make a global survey of the vertical structure of clouds and aerosols and their physical properties and, thus, provides a unique opportunity for GCM cloud evaluation. In the present study, the CALIPSO Level 2 Cloud Layer Product for an entire year period (June 2006 to May 2007) is used to analyze global distribution of cloud amount with a focus on the seasonal variations of cloud distribution. The observations are compared to results from global simulations performed using a few GCMs. The GCMs are used at the National Climate Center of China, the National Meteorology Center of China, and the Institute of Atmospheric Physics, Chinese Academy of Sciences, respectively, for operational prediction of weather or climate and climate study. Possible causes for the discrepancies in cloud distribution between the GCMs and CALIPSO are discussed and related to uncertainties associated with the GCMs' physics. Comparisons of atmospheric temperature and moisture and grid-scale vertical velocity are also made among the GCMs in order to explain the differences and similarities in their simulated cloud amounts.
A41B-0427
Selection of Data for Langley Derived Top Atmosphere Reference Values at a Turbid and Humid Coastal Site.
The motivation for this work is to preform in situ calibrations of a Multi Filter Rotating Shadow-band Radiometer (MFRSR) at a turbid and humid ocean site, thus eliminating the need to remove it from service for calibration. In pursuit of this goal, a method of selecting data used to preform Langley top of atmosphere (TOA) extrapolations, with relatively constant total optical depth (TOD) is presented. The TOA values, in this case voltages, can be used to derive atmospheric extinction values, and thus atmospheric column densities of various constituents. The data used is this study were collected at the Mauna Loa Observatory (MLO), Hawaii and at the Clouds and the Earth's Radiant Energy System (CERES) Ocean Validation Experiment (COVE) site located 20km off the Virginia, USA coast. MLO is an ideal site for this kind of work, hence the data collected there are used as a reference data set. A set of selection criteria, which select for Langley extrapolation periods, of constant TOD, are applied to the data. The resulting TOA values obtained at COVE are compared to those for MLO. The data were collected using a single MFRSR deployed first at MLO, then at COVE, again at MLO and, finally COVE. The COVE data are shown to have merit for the 500, 615, 673, and 870nm wavelengths but not for the 415 and 940nm wavelengths. Depending on the wavelength, differences in TOA voltages can be as little as 1% (673nm channel) to as much as 30% (940nm channel). This method can be applied to other sun photometers.
A41B-0428
Parameterization of Shortwave and Longwave Radiative Properties of Ice Clouds for Use in Climate Models
Climate modeling requires that the parameterization of the radiative effects of ice clouds be as accurate as possible. In this study, new parameterization of shortwave and longwave radiative properties has been developed on the basis of simulated single-scattering properties of nonspherical ice crystals and in-situ measurements of the microphysical properties of ice clouds observed during several field campaigns. Since the computation of the solar or infrared spectral radiation by accumulating individual bands over the whole spectral is computational costly, a common efficient method to calculate spectral radiation is based on several broad bands. Six parameterization schemes using different broad bands have been performed for shortwave and longwave radiative properties of ice clouds. The computed ice cloud radiative forcing at the top of atmosphere and Earth surface and heating rate from the parameterization schemes have been compared. It is found that the results from the different parameterizations agree well with each other. The relative errors of the results are generally less than 3%. The present parameterizations have been validated by comparing the ice cloud radiative forcing inferred from the parameterization schemes with those without parameterizing. The maximum error is about 5%. Moreover, the parameterizations have been applied to investigate the sensitivity of the radiative forcing and heating rate to the microphysical, macrophysical, and optical properties of ice clouds.
A41B-0429
Study of Light Scattering and Reflectance by an Atmosphere with Nonspherical Mineral Dust Aerosols
The scattering properties of dust aerosols assumed to be a mixture of randomly oriented spheroids with size and shape distributions are simulated with a composite method based on the T-matrix method and the geometric optics method (GOM). The T-matrix method is used only for small and moderate particles, whereas the GOM is applicable to large particles. The performance of this composite method used for the simulation of scattering properties of nonspherical dust aerosols is tested and evaluated based on measured scattering matrix of feldspar particles. The comparison between simulated and measurement indicates that use of spheroids with size and shape distributions can be a reasonable approach for estimating the scattering properties of nonspherical dust aerosols. Furthermore, Reflectance by an atmosphere containing nonspherical dust aerosols based on a polarized radiative transfer model is calculated, and sensitivity studies with respect to aerosol optical depth and particle sizes are carried out. Reflectance differences between vector and scalar radiative transfer models under different conditions are also discussed.
A41B-0430
Modeling of Single Scattering by Inhomogeneous Hexagonal Ice Crystals
Spherical or spheroidal air bubbles are generally trapped before escaping from rapidly growing ice crystals. The intent of this effort is to study the effect of the number, shape, size and location of the air bubbles inside hexagonal ice crystals on the single-scattering properties of these ice crystals. A theoretical model based on the geometric-optics method has been developed to simulate the scattering of light by randomly oriented ice crystals in the form of hexagonal ice columns containing spherical or spheroidal air bubbles. In this model, a combination of the ray-tracing technique and the Monte Carlo method is used. Our preliminary results indicate that the air bubbles inside ice crystals may lead to smooth phase functions and decrease the 22° and 46° halo peaks. Moreover, the backscattering is substantially reduced for the inhomogeneous particles, which physically is because of the divergence of light beams after their interactions with the air bubbles. These features become more pronounced with increasing the number and sizes of the air bubbles.
A41B-0431
ABI Solar Channel Bidirectional Reflectance Simulation
The objective of this study is to develop an ABI simulator (specifically, a fast radiative transfer model for single- layer clouds). The first step of this effort is to develop the lookup tables of the bidirectional reflectance of ice clouds at corresponding ABI channels. To develop the ice cloud bidirectional reflectance lookup tables, we first recomputed the shortwave single- scattering properties of ice crystals with geometries of droxtals, hexagonal plates, hollow columns, solid columns, 3D bullet rosettes and aggregates. The single-scattering properties were computed for 45 size bins ranging from 2 to 9500 ?m. This scattering database contains the asymmetry factor, single-scattering albedo, extinction cross-section, and scattering phase function. The data of individual ice cloud particle size distributions are derived from five field campaigns: a) FIRE-I; b) FIRE-II; c) ARM-IOP; d) TRMM, and e) CRYSTALFACE. Furthermore, the bulk scattering properties are averaged with a final set of 9 De values ranging from 5 to 200 ?m with an increment of log(De/2)=0.2. Each De model includes the microphysical and scattering properties such as ice water content, extinction efficiency, single-scattering albedo, asymmetry factor and scattering phase function. Integrating the bulk scattering properties with a discrete ordinate radiative transfer model (DISORT), we calculated the lookup tables of the bidirectional reflectance of ice clouds at corresponding ABI channels. Furthermore, we also compared the radiances simulated at the ABI and MODIS channels.
A41B-0432
Scattering properties of horizontally oriented hexagonal plates
Some ice crystals (ice plates, in particular) may be quasi-horizontally oriented in the atmosphere. For the backscattering of such oriented nonspherical particles, the conventional ray-tracing method is not applicable because the phase function derived from the ray-tracing calculation in this case is just the superposition of the diffraction contribution and two delta functions. For the scattering particles involved in this study, the applicability of rigorous methods such as the T-Matrix, DDA, FDTD/PSTD methods are also limited because the typical sizes of ice crystals are a few hundred microns and the corresponding size parameters are too big for these methods. To illustrate the inapplicability of the conventional ray-tracing method to the scattering of light by horizontally oriented ice plates, we derive the internal scattered-field inside these particles on the basis of the electromagnetic wave theory by ignoring the effect of side faces, which is then used to derive the phase matrix. It is shown that the phase function derived from the electromagnetic wave theory is quite different from the counterpart from the ray-tracing method.
A41B-0433
Investigating the Shortwave Radiative Effects of Cloud Field Geometry in the Tropical Western Pacific
In the last decade numerous advances have been made in parameterizing cloud-radiation interactions. Recent improvements have involved statistical representations of the influence of three-dimensional cloud fields on the domain averaged radiation. A stand-alone comparison of traditional plane-parallel shortwave radiative transfer code against a statistical algorithm and observations is performed using four years of data from the Atmospheric Radiation Measurement Program's Tropical Western Pacific Clouds and Radiation Testbed Site. Statistical cloud properties are derived from observed cloud chord lengths and input into the stochastic model, removing the necessity of explicitly simulating individual cloud fields and allowing for extended model runs. The purpose of the comparison is to identify objectively meteorological situations where 3-dimensional cloud field geometries are radiatively significant such that a statistical approach to radiative transfer, such as the stochastic technique, is more appropriate than the traditional plane-parallel approach. Results suggest that the relationship between vertical extent of liquid cloud and liquid water path is one indicator of radiatively important cloud regimes, as is the relationship between solar zenith angle and cloud fraction. A simple-parameterization based on criteria developed from these results is applied to the plane-parallel model to represent the effect of complex cloud field geometry. The preliminary runs show improvement in the performance of the plane-parallel model when compared to observations.
A41B-0434
High Spectral Resolution Lidar Observations of Diamond Dust Layers in Eureka, Canada.
Surface-based ice crystal layers, also referred to as diamond dust layers, occur frequently during the dark season in the Canadian High Arctic. They form under cold winter conditions (temperatures below 260K) in the stable boundary layer. With an average height of 400m, as calculated from a data set spanning the winter months of 2006, they are typically decoupled from higher cloud features, but can also be capped by a thin layer of supercooled water. In the present work, diamond dust layers are observed with the University of Wisconsin Arctic High Spectral Resolution Lidar (AHSRL) based in Eureka (79.99N, 86.93W), in the Nunavut Territory. The different ice crystal layers are characterized in terms of backscatter cross section and depolarization ratio. Large variability in linear depolarization ratio within single or multiple diamond dust events is observed. Possible causes are variations in particle size, particle shape and orientation, and the presence of aerosols and/or liquid water within the diamond dust layer's volume. The latter possibility is investigated as part of the Canadian Network for the Detection of Atmospheric Change (CANDAC). We expect the results to improve the understanding of diamond dust formation processes and physical structure. These have numerous implications for boundary layer processes, such as radiative transfer, moisture exchanges and pollution events.
A41B-0435
Retrieval of Aerosol Optical Thickness and Normalized Water-Leaving Radiances From the SeaWiFS and MODIS Sensors Over the Chesapeake Bay Area (Case 2 Water)
Presently, a suite of 12 aerosol models are used for atmospheric correction purposes to retrieve normalized water-leaving radiances in the visible bands of the SeaWiFS and MODIS sensors. These aerosol models are based on Shettle and Fenn's models (1979) of tropospheric and oceanic aerosols. Over most of the open oceans of the world (case 1 water), the atmospheric correction algorithm has been shown to work reasonably well. However, over case 2 waters, (for example Chesapeake Bay) the algorithm often yields negative water- leaving radiances, particularly, in the blue bands of the two sensors. In addition, over the coastal areas, the retrieved aerosol optical thickness (AOT) in the 865/869 bands are often higher than the in situ AERONET retrievals. Our analysis of the AERONET data show that Shettle and Fenn's aerosol models are not representative of the aerosols generally found over the coastal region of the Eastern United States. We show that use of wrong aerosol models often results in negative water-leaving radiances. Also, the backscattering of the solar irradiance in the near IR bands by phytoplankton and non-algal suspended particles results in overestimation of AOT. Based on the AERONET data, we have developed a set of new aerosol models for the atmospheric correction over Chesapeake Bay. Results from the new aerosol models, including comparison of satellite-derived AOT and the AERONET in the visible and near IR bands, will be presented.
A41B-0436
Relation Between Backscatter and Depolarization Ratio for ISCCP Cloud Types On the Basis of Collocated MODIS and CALIPSO products
Using the International Satellite Cloud Climatology Project (ISCCP) cloud classification, we investigated the relation of backscatter and backscattering depolarization ratio for nine types of clouds on the basis of the collocated data of the cloud products derived from the measurements made by the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) and the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard Aqua. Specifically, MODIS MYD06 cloud optical thickness and cloud top pressure are used to classify cloud types. For each cloud type, layer-averaged backscatter and backscattering depolarization ratio from the CALIPSO measurements are investigated. High clouds (cirrus, cirrostratus and deeply convective clouds) and mid- and high-latitude altostratus clouds show signatures of both ice and water clouds. We also use MODIS cloud phase flags to screen ice clouds. As a result, the classified high clouds flagged as water clouds show only water phase feature; however, the ice counterparts still show features of both ice and water phases. To understand the depolarization ratio of ice clouds, we also simulated the depolarization ratio for ice crystals using an improved geometric optics model. It is shown that the linear depolarization ratio of ice particles depends on particle habits, which however is not sensitive to effective particle size. Furthermore, we compared the values of the backscattering linear depolarization ratio computed with various habit and size distributions.
A41B-0437
The Hybrid Matrix Operator - Monte Carlo Method for Solving the 3D Vector RTE in Atmosphere-Ocean Systems
An impulse response solution (Green matrix) is developed to solve the vector radiative transfer equation (VRTE) in a 3D atmosphere-ocean system (AOS). The system is divided into three parts: the atmosphere, the dielectric interface, and the ocean. For the atmosphere and ocean, the impulse response functions are calculated by the Monte Carlo method. For the dielectric interface, the impulse response function is calculated by the Fresnel formulas. The matrix operator method is then used to couple these impulse response functions to obtain the vector radiation field for the AOS. The primary advantage of this hybrid method is that it solves the VRTE efficiently in an AOS with a time dependent dielectric interfaces but with the same atmospheric and oceanic conditions. We present the downward radiance field in an ocean with a real time dependent sinusoidal ocean wave.
A41B-0438
The Nature of the Radiance and Polarization in Deep Oceans
We studied the asymptotic nature of the radiance and polarization in the underwater light field in an atmosphere- ocean system. We used a vector radiative transfer code to solve for the total Stokes vector and found that the radiance of the light field becomes asymptotic more quickly than the polarization does. Our simulations imply that for a homogeneous water body described by a Petzold phase function, a single scattering albedo of ω_0 ≥ 0.8 is required to reach the asymptotic regime practically, otherwise, the radiance in the asymptotic regime becomes too small to be detected. For the asymptotic regime to be realized physically, a water body with a less anisotropic phase function and/or a larger ω0 is necessary. For a real water body described by a Case 1 water model, the asymptotic regime could be reached at wavelengths 400 nm <λ< 500 nm. The effects of Raman scattering have also been included, and it turns out that for wavelengths λ < 540 nm, the contribution from Raman scattering can be reasonably neglected in the study of the asymptotic radiance.
A41B-0439
Analytic Expressions for Aerosol Light-Scattering Cross Section and Angstrom Exponent
The light-scattering cross section σsca and its dependence on wavelength λ, which is parameterized by the Ångström exponent å, are intrinsic properties of an aerosol. These quantities are important for understanding Earth's radiative balance and yield information on the sizes of particles that provide the dominant contribution to the scattering; aerosols for which the scattering is dominated by particles much smaller than the wavelength are characterized by larger values of å than aerosols for which scattering is dominated by much larger particles, for which å is near zero. For an aerosol consisting of spherical particles σsca can be calculated by integration over the size distribution given knowledge of the scattering efficiency of a particle Qsca, which depends on the radius of the particle, the wavelength, and the index of refraction m, and å can be determined from the derivative of σsca with respect to wavelength. However, calculation of these quantities in this manner is computationally laborious and does not explicitly illustrate the dependences on properties of the size distribution or on wavelength or relative humidity. By using scaling arguments and an approximation for the scattering efficiency, an analytic expression can be derived for the scattering cross section of an aerosol with a lognormal size distribution characterized by geometric radius r0 and geometric standard deviation σ that is accurate over a wide range of these parameters. This expression is computationally much more efficient, and permits derivation of a simple analytic expression for the Ångström exponent å=-1.14-2(lnσ)2-ln[r0(m-1)/λ]/0.28+(lnσ)2 which explicitly shows the dependence on parameters of the size distribution r0 and σ and on the wavelength and index of refraction. Other formulae are presented for situations outside the range of validity of this expression, permitting a computationally efficient yet accurate means of determining the Ångström exponent.
A41B-0440
Light Scattering by Nonspherical ice Crystals: Comparison of Pseudo-spectral Time Domain and Discrete Dipole Approximation Methods
We apply the pseudo-spectral time domain (PSTD) and discrete dipole approximation (DDA) methods for the solution to the scattering by randomly oriented ice crystals (hexagonal columns, plate, droxtals, bullet rosettes, and aggregates). The optical properties (extinction efficiency, single-scattering albedo, asymmetry factor and phase matrix) of these particles are calculated in the cases of small and moderate size parameters at wavelengths of 0.66 μ m and 12 μ m. The computational efficiency of these methods in terms of their demand on memory and CPU time is contrasted and analyzed. The configuration of particle random orientations in this study is specified with the help from the T-matrix method in the cases of spheroids and cylinders.
A41B-0441
Investigation of the Radiative Forcings of Thin Cirrus in the Tropical Atmospheres Based on AIRS/ARM Data
We developed a fast thermal infrared radiative transfer model (Yue et al 2007) to retrieve thin cirrus clouds using Atmospheric Infrared Sounder (AIRS) data. The delta-four-stream approximation has been incorporated into our model to account for multiple scattering contributions in the infrared cloudy spectra, important for cirrus with optical depths larger than about 0.3. We apply this improved retrieval method to a number of nighttime thin cirrus scenes in the tropics selected from the AIRS dataset to determine cirrus optical depth and ice crystal size and habit. The Fu-Liou broadband radiative transfer program has been modified to include observed ice particle habits in the parameterization. The solar and infrared radiative forcings and heating rates produced by thin cirrus in the tropical atmosphere have been analyzed using the retrieved cirrus properties along with the new parameterization. Validation of the cirrus retrieval and the computed broadband fluxes has been carried out by comparing them with the ground-based measurements available from the Atmospheric Radiation Measurement (ARM) program. Finally, we will report investigation of the effect of thin cirrus clouds on the radiation budget at the top of atmosphere, at the position of the tropical tropopause, and at the surface.
A41B-0442
Physical Models of Surface Polarization
Accurate models of polarized surface reflectance are essential for atmospheric aerosol retrieval from remote sensing polarimeters such as the Aerosol Polarimetry Sensor (APS) on the upcoming NASA Glory mission. However, most existing models are ad hoc modifications of the Fresnel reflection coefficient. These modifications require scale factors that are large and unrelated to the ad hoc justification for the use of the Fresnel coefficient. In addition, measurements from recent field campaigns show polarized reflectance in the backscatter that cannot be reproduced by polarized surface reflectance models that use the Fresnel coefficient. Here we present a physically based polarized surface model. We use geometric optics models of large plates and needles to describe leaves and spheroidal particles for grains of soil. These are compared to the measured polarization of various surfaces and the efficacy of these physical models are evaluated.
A41B-0443
The behaviour of cloud and clear sky brightness in the vicinity of the cloud edge
L.P. Bass1, O.V. Nikolaeva1, V.S.Kuznetsov2, A. A. Kokhanovsky3,4 1Keldysh Institute of Applied Mathematics, Russian Academy of Science Miusskaya Sq. 4,125047 Moscow, Russia 2Research Scientific Center "Kurchatov Institute", Kurchatov Sq. 1, 123182 Moscow, Russia 3Institute of Remote Sensing, Bremen University, Otto Hahn Allee 1 28334 Bremen, Germany 4Institute of Physics, National Academy of Sciences of Belarus, Nezaleznasti Pr. 70 220072 Minsk, Belarus In the solution of remote sensing problems in the framework of the Independent Pixel Approximation (IPA) the horizontal transport of radiation is not taken into account. Therefore, the large errors in the retrieved optical parameters of a medium under study can occur in retrievals for regions, where 3-D radiative transfer effects are of importance (Wen et al, 2007, Titov, 1998). In the present work we analyze the brightness at the edge of a cubic cloud. The energy balance equations within the clear sky-cloud boundary layer are studied. The boundary layer is the domain that includes the vertical boundary of the adjacent pixels with different optical properties. Balance equation connects the radiation fluxes entering into the boundary layer and outgoing from it, and also the amount of energy absorbed in the layer. It is demonstrated that horizontal transport of radiation generates several observable phenomena such as "shadowing" and "brightening" (depending on the Sun position with respect to the cloud and also the area studied). All calculations are performed with the code Raduga-5.1 (Nikolaeva et al., 2005) developed for the computer with the parallel architecture for 1-D, 2-D, 3-D radiative transfer. The code is based on the numerical solution of the integro – differential radiative transfer equation (RTE) with correspondent boundary conditions and prescribed properties of a light scattering medium. Grids with respect to spatial and angular variables are introduced and RTE reduced to the system of the grid equations. The derived system of equations is solved using standard finite difference techniques. The comparison with Monte-Carlo (see www.libradtran.org) calculations demonstrated a high accuracy of the method. The differences between these two completely different types of 3-D RTE solution are within 1%. References Nikolaeva, O. V., L.P.Bass, T.A.Germogenova, A.A.Kokhanovsky, V.S.Kuznetsov, B. Mayer, 2005: The influence of neighbouring clouds on the clear sky reflectance studied with the 3–D transport code RADUGA, J. Quant. Spectr. Rad. Transfer, 94,405-424. Titov, G. A., 1998: Radiative horizontal transport and absorption in stratocumulus clouds, J. Atmos. Sci., 55, 2549- 2560. Wen, G., A. Marshak, R.F. Cahalan, L.A. Remer, and R.G. Kleidman, 2007: 3-D aerosol-cloud radiative interaction observed in collocated MODIS and ASTER images of cumulus cloud fields. J. Geophys. Res., 112, D13204, doi: 10.1029/2006JD008267.
A41B-0444
Terahertz Remote Sensing of Ice Clouds - Sensitivity on Ice Dielectric Properties
Initiated by current developments in terahertz sensor technology the application of instruments operating in the spectral region between 0.1 - 30 THz is considered for a number of remote sensing issues. Accounting for more than 50 percent of the outgoing longwave radiation and with the major component of cirrus radiative forcing in the far-infrared, satellite measurements in this spectral region will significantly support the determination of the radiation budget of the Earth. Furthermore, spanning the whole range of particle sizes found in tropospheric ice clouds, the Terahertz region bears the potential to complement existing methods and improve our knowlegde and understanding of those clouds. Both, determination of the Earth's radiation budget as well as retrieving ice cloud properties require appropriately accurate calculations of radiative transfer. Hence, a good knowledge of the input parameters to the radiative transfer models is needed. In particular, this includes spectrally dependent properties of the molecular as well as particulate atmospheric matter, i.e., spectroscopic parameters of the molecular absorption lines and continua as well as the dielectric properties of aerosol and cloud particle material. Due to the lack of Terahertz light source and receiver technology in the past, measurements of these parameters have been sparse and the knowledge about them is rather poor. In preparation to evaluate the feasibility of monitoring tropospheric ice clouds using passive Terahertz observations, we study the modeling uncertainties due to the unconfident knowledge of the complex refractive index of ice. We give an overview of the consistency and discrepancies, respectively, of the existing measurements and models for ice refractive index in the Terahertz region. Using calculations of particle optical properties according to Mie theory as well as the radiative transfer models Moliere and SARTre, we estimate the deviations in particle optical properties and simulated observation spectra arising from applying the different ice refractive index data. Furthermore, in order to derive the uncertainty in retrieved cloud properties the sensitivity of the atmospheric spectra to the ice refractive index on the one hand and cloud ice content as well as particle size on the other hand is compared. From that, conclusions will be drawn on requirements for future experiments to measure dielectric properties of ice for geophysical applications.
A41B-0445
Numerical solution of single-scattering properties of dust
Satellite or airborne measurements show significant difference between dust aerosols and cloud particles in light scattering properties. The modeling of single scattering properties of dust particles is important for remote sensing of regional or global climate. In this work, light scattering by irregularly-shaped dust particles is studied with the finite-difference time domain technique. For given Gaussian particle shapes and size parameters in the resonance region, the scattering phase matrices and asymmetry factors are calculated. It is found that the deformation of the particle surface can significantly smooth the scattering phase functions. The polarization properties of the scattered light by irregular dust particles are also significantly different from those by spherical particles. Based on the polarization properties of scattered light from dust aerosols, a remote sensing method which could accurately measure the shapes of the dust particles is proposed.
A41B-0446
Linearization of Monte Carlo vector radiative transfer model MCC++ for aerosol retrieval
Linearized radiative transfer models are characterized by capability to calculate directly and efficiently the weighting functions used in retrieval algorithms, but not only radiance. Their application to interpretation of observations improves accuracy of retrieval of atmospheric gases, properties of aerosol and clouds due to proper calculation of multiple scattering. The model MCC++ describes transfer of the polarized light in the spherical-shell atmosphere. The model MCC++ was previously linearized to calculate derivatives with respect to the volume absorption coefficient, the Lambertian and BRDF surface properties. A new version of the model was linearized to provide derivatives of four Stokes parameters with respect to the aerosol volume scattering coefficient. The developed method of calculation allows insignificantly increase time of calculation in comparison with calculation of four Stokes parameters only. The deduced representations of derivatives improve ability of the RT model to investigation of the aerosol retrieval capability of different types of satellite measurements. Formalizm for Monte Carlo calculation of the derivatives and examples of numerically simulated aerosol measurements are presented. http://postylyakov.narod.ru/
A41B-0447
3D Thermal Infrared Radiative Transfer in Mountains
We developed a 3D Monte Carlo photon tracing program for radiative transfer in inhomogeneous and irregular terrain coupled with the correlated k-distribution method for gaseous absorption in the atmosphere for the calculation of broadband thermal infrared (IR) fluxes at mountain surfaces. The thermal IR radiative transfer program includes emission from the atmosphere to the surface and vice versa as well as emissions between mountain surfaces. Both the atmosphere and the land surface are discretized by using finite cubic cells characterized by the spectral optical properties of molecules and background aerosols (absorption coefficient, single-scattering albedo, and scattering phase function) and terrain configuration (albedo, elevation, slope, and orientation). The emissivity of gases is parameterized in terms of the vertical optical depth of cubic cell. We selected an area of 100×100 km2 in the Tibetan Plateau near Lhasa city with a horizontal resolution of 1 km2 and used the surface temperature and albedo available from MODIS/Terra dataset for this study. We show that surface temperature is the dominating factor in radiative transfer calculations and that subgrid variability of the net surface IR flux distribution relative to a flat surface (1D) with average elevation and temperature can be as large as 50 W/m2 at cold mountain surfaces.
A41B-0448
Detection and Retrieval of Mineral Dust Aerosols Using AERI Data
A dust detection/retrieval method using ground-based Atmospheric Emitted Radiance Interferometer (AERI) brightness temperature spectra has been developed. Taking advantage of the high spectral resolution of AERI, we exploit differences between the spectral absorptive power for dust and cloud in prescribed thermal IR window sub-bands to separate dust from clouds, and to retrieve dust IR optical depths. Dust composition was prescribed using the refractive index datasets for minerals commonly observed around the United Arab Emirates (UAE) region including quartz, kaolinite, kaolinite mixed with hematite and calcium carbonate and for comparison with the individual minerals, the refractive indices of the Volz Saharan dust model were also examined. Five dust microphysical models were constructed using in-situ data from the UAE Unified Aerosol Experiment (UAE2). The single-scattering properties for oblate spheroids and hexagonal plates, two particle geometries routinely interpreted in electron microscopy, were computed using the T-matrix and FDTD programs. Sensitivity of the AERI spectra to dust composition, shape, size, and precipitable water vapor was investigated using the CHARTS radiative transfer program. AERI data for four dust and cirrus cases from the UAE2 field campaign were selected to demonstrate the effectiveness of the detection/retrieval approach. The AERI detection/retrieval results compare well with the values determined from coincident and collocated MPLNET micro pulse lidar and AERONET sun-photometer measurements respectively. The AERI retrieved optical depths (scaled from IR to visible) were found to be within 30% of those measured by AERONET. This novel AERI detection/retrieval approach will help detect and track regional dust events, and quantify dust IR radiative forcing parameters for both daytime and nighttime conditions.
A41B-0449
Scattering Theory for Lidar Remote Sensing Applications: Progress and Challenge
Cloud, aerosol and ocean studies using CALIPSO data are based on our understanding of single and multiple scattering of various particulates and ocean surface. In this presentation, we discuss the applications of scattering theory and model in CALIPSO data analysis, and the areas that need more theoretical studies. This includes, 1. 532nm calibration. CALIPSO lidar uses molecular backscatter for calibration. Our discussion includes, A. What is the physics basis of this calibration procedure and its uncertainty? B. How do we assess the calibration performance using other well known targets, such as water clouds? 2. 1064nm calibration. CALIPSO uses thick ice clouds for 1064nm calibration while assuming constant lidar ratio for these clouds. Theoretical studies are needed to answer these questions, A. Are the color ratios for thick ice clouds constant? B. Can we use other objects (water clouds, ocean surface) to verify that? 3. Multiple scattering of clouds and aerosols. A. How do we estimate multiple scattering factors of ice clouds, water clouds and aerosols? B. How do we separate multiple scattering from detector transient response in CALIPSO data? 4. Cloud phase studies, A. How does CALIPSO identify cloud phase? B. What is the signature of horizontally oriented ice particles? 5. Ocean studies, A. Can we see ocean sub-surface from CALIPSO 532nm perpendicular channel? B. How accurate can we derive ocean surface winds and atmospheric column optical depths? 6. Long term stability, A. Is CALIPSO molecular calibration changing from month to month? B. How well can CALIPSO detect cloud and aerosol trends? We will present where we are in terms of understanding these questions.