Clouds have always been an obstacle to remote sensing using visible and infrared wavelengths. Visible and infrared measurements are altered or obscured by their presence. Clouds also are one of the controls of the heat balance of the earth. Of the different forms of clouds, cirrus have been the most elusive to detect. Ground observers under report them because they are often obscured from view by lower clouds and not considered important to aviation. Satellite measurements also have trouble detecting cirrus because their transparency makes them difficult to distinguish from ground or lower cloud backgrounds. Satellite measurements are altered by the presence of these semitransparent clouds and any remote sensing system will have to account for them or use extensive methods to detect and avoid them. The high frequency of cirrus clouds over the earth was not realized until recent sateffite cloud data sets were analyzed. There have been only four clouds studies that have covered the entire earth and have included a long enough period to be used for population statistics. They are the summary of surface weather observers by Warren et al. (1988), the limb scanning Stratospheric Aerosol and Gas Experiment (SAGE see Woodbury and McCormick, 1993)) of NASA, the International Satellite Cloud Climatology Project (ISCCP see Rossow and Lacis, 1990) and the analysis of HIRS data by Wylie et al. (1994). The frequency of semi-transparent cirrus in these studies varies from 10-45%. This reported amount of drrus varies among the studies because of differences in their ability to depict semi-transparent clouds and distinguish them from other land and ocean backgrounds or other clouds below them. Very thin cirrus are often missed by these cloud detection schemes and thus the frequency of thin cirrus may be even higher than reported by any of the studies.

As part of an overall plan for upgrading its existing real time cloud analysis (RTNEPH) and forecast system the Air Force has sponsored the Support of Environmental Requirements for Cloud Analysis and Archive (SERCAA) initiative. The main focus of SERCAA is development of a cloud analysis model capable of analyzing multisource satellite data from NOAA/AVHRR, DMSP/OLS and the multiple geostationary platforms. Separate cloud analysis algorithms have been developed which exploit the inherent strengths of each sensor suite, with the multiple analysis results then integrated to produce a single consistent global analysis. The cloud analysis parameters which will be merged include total cloud fraction, number of cloud layers (up to 4 `floating' layers will be permitted), layer cloud fraction, layer height, and layer type. The integrated analyses are generated on a polar stereographic grid at a spatial resolution of approximately 25 X 25 km. Test case results using data from 1993 are presented which highlight various aspects and of the analysis and integration algorithms.

The Cloud Depiction and Forecasting System II (CDFS II) is a major new initiative that will transition the Air Force Global Weather Central (AFGWC) to a new satellite data processing system and include extensive changes in cloud analysis/forecasting at AFGWC. The present cloud analysis model, the RTNEPH, combines reduced resolution DMSP OLS or NOAA AVHRR data with conventional observations. The RTNEPH domain and analysis frequency are limited by its dependence on polar-orbiting satellites. In the CDFS II era (1998+), AFGWC cloud forecast models will benefit directly from improved automated nephanalysis capabilities from multiplatform sensor data. Support of Environmental Requirements for Cloud Analysis and Archive (SERCAA) project is a research and development program sponsored by the Strategic Environmental Research and Development Program that will provide both the next generation nephanalysis model for CDFS II and a new global cloud algorithm for use in determining the radiative and hydrological effects of clouds on climate and global change. SERCAA cloud analysis products will be available to a wide community of users both within and outside of the Department of Defense. The SERCAA project consists of two phases, the first has provided algorithms for retrieval of cloud spatial parameters for the CDFS II initiative. The second phase is concentrating on development of radiative and microphysical cloud parameter algorithms and on the archive structure.

Algorithms used to produce automated cloud analyses from global meteorological satellite imagery must compensate for variations in cloud spectral signatures caused by changes in atmospheric attenuation and solar scattering geometry that occur as the satellite orbits the Earth. In this paper, a methodology is presented that describes the variations in the spectral signatures of optically-thick (water) clouds for a wide range of solar illumination conditions. Functions are developed that describe these changes and are demonstrated in the analysis of high resolution NOAA AVHRR imagery. The accuracy of the automated cloud analyses is measured against ground truth (manual) cloud, no-cloud analyses.

The performance of an algorithm which identifies and removes clouds from ocean color satellite data over the ocean is evaluated. The evaluations were accomplished with archived sets of Coastal Zone Color Scanner (CZCS) and simulated SeaWiFS data. The simulated SeaWiFS sets were produced by spectrally and spatially convoluting the data of the Airborne Visible-Infrared Imaging Spectrometer to fit the spectral responses of the SeaWiFS channels and achieve the same spatial resolution of the SeaWiFS instrument. This study also included evaluations of the effect that changes in spatial resolution have on the texture computations of the algorithm. Convolution procedures were used to generate images at various spatial resolutions: 20 m, 50 m, 100 m, 500 m and 1 km. Results from the investigations involving CZCS Channel 5 (720 - 800 nm) data indicate that the algorithm performs well in all the oceanic data sets tested but not in the coastal imagery. Cloud masking in the coastal sets was clearly affected by bottom reflectances and river run off. In the evaluations involving simulated SeaWiFS Channel 8 (845 - 885 nm) data, the algorithm performed accurately every time and was not affected by coastal or oceanic features. It began to confuse cloud shadow edges with cloud edges at resolutions of 30 m or less. From the results obtained, it appears that this algorithm has the potential to identify clouds not only in SeaWiFS data, but also in data from high resolution color sensors.

Cloud detection from satellite data usually relies on the assignment of reflectance and radiation temperature of a pixel to a cloud or a cloud free surface according to preset thresholds. It also exploits the contrast between adjacent pixels. It will be shown that the so called split window information like that provided by the `Advanced Very High Resolution Radiometer' (AVHRR) onboard the NOAA satellites is a powerful tool to reduce ambiguity in cloud detection. The reflected part of the medium infrared spectral information is used for phase detection. Double view information like that of the `Along Track Scanning Radiometer' (ATSR) on ERS-1 may be of substantial help in cloud identification, but a careful georeferencing is required. An algorithm package called `AVHRR Processing scheme Over cLouds, Land and Ocean' (APOLLO) which can also process ATSR data, is used to derive cloud masks from AVHRR and ATSR data for day and night satellite overpasses. Several products can be obtained from the image files together with the cloud mask. From the cloud free pixels surface products can be derived. Cloud cover is derived from all pixels while cloud properties like the liquid water path are derived from fully cloudy pixels only. Examples are shown which are derived from a regional cloud climatology.

Bispectral techniques employing the infrared (IR: 10.5 - 12.5 micrometers ) and water vapor (WV: 5.7 - 7.1 micrometers ) channels of the geostationary Meteosat satellites are in use for cloud classification and cloud altitude correction of semi-transparent cirrus clouds. This paper investigates the relationship between radiance observation in the IR and WV channel and the validity of various assumptions. Specifically the following problems are addressed: (1) accuracy of coregistration of two simultaneous images in the IR and WV channel, (2) the effect of instrument noise on single pixel classification, (3) the influence of moisture above the cloud top on the satellite observed radiances and (4) the effects of different cloud emissivities in both channels. The study makes use of both radiation model calculations and satellite observations. The impact of different ice particle distributions on cloud emissivity is studied.

Clouds play an important role in the hydrological cycle and they are a climatic component. It is therefore necessary to known the distribution of clouds. A high-spatial resolution climatology of monthly mean cloud cover has been produced from the daily afternoon passages of the NOAA satellites since September 1991. The APOLLO (AVHRR Processing scheme Over cLouds, Land and Ocean) algorithm has been applied to AVHRR data to obtain a pixel by pixel cloud mask for an area of 560 X 560 km² over the Alps and their forelands. Cloud cover is calculated for subareas of 14 X 15 km² for the following classifications: all clouds, thick clouds, thin clouds, thick clouds with low, medium or high tops. This regional cloud climatology reveals topometeorological features like an increased cloud cover in the luff regions at the edges of the Alps. A strong horizontal inhomogeneity on a regional scale is typical in summertime. The seasonal changes of cloud cover are greater in the south than in the north of the Alps due to different climate types. Although the APOLLO algorithm still needs improvement it has proved to be a powerful tool for cloud detection and classification.

A cloud observational network is used to characterize cloud fields over a 120*120km^2 area in the Netherlands. The system consists of a network of stations for ground based remote sensing and a processing environment for AVHRR and Meteosat measurements. The aim is to obtain a detailed quantitative description of cloud properties within a gridbox of a climate model. Results are to be used for the improvement of parameterizations of clouds and cloud- radiation interaction in weather and climate models. Research will focus on the characterization of the variability in time and space of cloud properties. Currently the system is fully operational. Case studies are performed to investigate the individual instrument response to clouds and their correlation to other measurements. In this paper some results for 4 August, 1993 are presented.

In this paper we discuss the application of the T-matrix approach to rigorously compute light scattering by polydisperse, randomly oriented nonspherical particles with sizes comparable to and larger than wavelengths of observation. First, we describe an efficient method for suppressing the numerical instability of the regular T-matrix approach in computations for spheroids and finite cylinders with size parameters larger than about 25. Second, we describe how the T-matrix approach can be used to rigorously compute the scattering of light by randomly oriented two-sphere aggregates (bispheres). Both methods are extremely efficient, are applicable to scatterers with equivalent-sphere size parameters exceeding 50, and are especially suitable in computations for polydisperse particles. We report results of numerical computations that demonstrate the capabilities of the two methods and are used to discuss the effects of particle nonsphericity on light scattering.

Possibilities of remote sensing of cirrus cloud ice mass content using spaceborne millimeter wavelength radar are analyzed. It is shown that measurements of radar reflectivity alone are generally insufficient to get unambiguous estimates of ice mass content. Ice mass content-radar reflectivity relationships, which are usually used for such estimates are sensitive to observational and meteorological conditions. Potential accuracy of these estimates can be significantly improved if independent information about characteristic cloud particle sizes is used.

The model of homogeneous spheroids were chosen to provide the detailed comparison of two popular solutions of the light scattering problem: T-Matrix Method and Discrete Dipole Approximation. The exact solution by the Separation of Variables Method were used as a standard giving the most accurate results. We have computed the scattering cross-sections of prolate and oblate spheroids with the refractive index m equals 1.3 and 2.5 at fixed orientation in a wide range of the aspect ratios and sizes. We found that: (1) the coincidence of the T- Matrix Method and the Separation of Variables Method is very good (> 6 - 10 digits) up to some boundary particle size; for larger particles the precision of T-Matrix results sharply drops; (2) the Discrete Dipole Approximation code gives the satisfactory results (the deviations from other methods less 5 - 10%) for large values of size and aspect ratio even if the number of dipoles is 1,000 - 1,500; the accuracy less than 1% may be obtained if the number of dipoles exceed 10,000 - 50,000; (3) the accuracy of the methods decrease with the growth of the parameter (tau) equals m (DOT) (2(pi) r_v/(lambda) ) (DOT) (a/b), where r_v is the radius of equivolume sphere, (lambda) the wavelength of incident radiation, a/b the aspect ratio. If a/b <EQ 4, the coincidence of the results with those of the Separation of Variables Method is within 1 - 3% for r approximately equals 8 - 16 (Discrete Dipole Approximation) and r approximately equals 50 - 65 (T-Matrix Method). For the particles with a/b >= 10, the Separation of Variables Method is preferable, if 2(pi) r_v/(lambda) >= 2 - 3.

The new solution of the problem of light scattering by coated spheroids was used to calculate the optical properties of prolate and oblate particles. The solution was obtained by the method of separation of variables for confocal spheroids. We consider the silicate core ice mantle particles and present the extinction cross-sections for prolate and oblate spheroids with the refractive indices m_core equals 1.7 + Oi, 1.7 + 0.1i and m_mantle equals 1.3, the aspect ratio (a/b)_mantle equals 2 and various volume ratios V_core/V_total. The results are plotted for different size parameters x_v equals 2(pi) r_v/(lambda) , where r_v is the radius of equivolume sphere and (lambda) is the wavelength of incident radiation. The main conclusions are: (a) if V_core/V_total equals 0.5, the optical properties of a core-mantle particle are determined mainly by its core: for prolate non-absorbing spheroids when x_v <EQ 4.5; for prolate absorbing spheroids when x_v <EQ 4.5 or x_v > 10, for oblate absorbing and non-absorbing spheroids when x_v <EQ 10. (b) the linear growth of the extinction cross-sections on the volume ratio V_core/V_total is obtained for prolate particles with x_v equals 1 and 2. (c) the non-linear increase of cross- sections is obtained for oblate particles with size parameters x_v equals 1 - 5. (d) the small imaginary part of the core refractive index m [Im(m) < 0.01] practically does not change the optical properties of an inhomogeneous particle. When the imaginary part reaches 0.1, the noticeable changes of cross-sections may be detected.

The solution of the electromagnetic scattering problem for cofocal coated spheroids obtained by the method of separation of variables in a spheroidal coordinate system is presented. The main features of the scheme of the solution are: (a) the incident, scattered and internal fields are divided into two parts: the axisymmetric part which does not depend on the azimuthal angle and the non-axisymmetric one; the diffraction problem is solved independently for each part; (b) the scalar potentials for the solution of each problem are chosen by a special way: the Abraham's potentials (for the axisymmetric part) and superposition of the potentials used for spheres and cylinders (for the non-axisymmetric part). Then, we derive the systems of linear algebraic equations in the simplest form and can investigate them analytically. Such a scheme allows to solve the light scattering problem for spheroids with arbitrary asphericity and has an advantage in comparison with other approaches, especially for large values of the aspect ratio: the computational time is being reduced in about ten times for small values of the aspect ratio (a/b equals 2) and in about hundred times for a/b equals 10 in comparison with the well-known solution of Asano and Yamamoto.

We report on a new application of the High Resolution Infrared Radiation Sounder (HIRS) to the global retrieval of sulphuric acid aerosol from the 15^th June 1991 eruption of Mt. Pinatubo (15.1 degree(s)N,120.4 degree(s)E). The aerosol is retrieved by measuring the emitted radiance using the HIRS channel centered at 8.3 micrometers which is particularly sensitive to the aerosol. Another channel at 12.55 micrometers , which is relatively insensitive to the aerosol, is used to remove other variations in the signal. The transmission at 8.3 micrometers is retrieved and related to the column density. We present an aerosol climatology which will include latitudinal distributions of column density and monthly global mass loadings. We also attempt to retrieve particle mode radius by combining optical depth measurement (Lidar) in the visible with a retrieved column density at 8.3 micrometers . We compare and contrast lidar column density measurements with column densities obtained using HIRS.

Mineral aerosol can be observed from satellite, with shortwave (VIS) imagery over oceanic surfaces and longwave (IR) imagery over land surfaces (especially for arid regions). Techniques appropriate to the VIS and IR channels of satellites of the current Meteosat series are described. Over ocean, the VIS technique involves a retrieval of aerosol optical depth obtained from the satellite-derived data. Over continent, the IR technique requires creating optimized reference images, constituted by maximum radiances associated with clear and clean pixels. Clouds and dust plumes are then separated from the permanent surface structures in difference images obtained by subtracting original images from the reference one. Both techniques are applied to data at 12 UTC in the format ISCCP B2. The respective patterns of retrieved aerosol optical depth over ocean and derived difference images over continent are observed to fit satisfactorily along the coast. Results of optical distribution and temporal variation of the Saharan aerosol occurrence over Africa and its bounding seas and oceans, north of the equator are presented. The seasonal character of the phenomenon is described. Observed maximum for these frequencies allows the sources of dust emission to be located and their seasonal rhythm of activity to be estimated.

Atmospheric aerosol particles play a vital role in the earth's radiative energy budget. They exert a net cooling influence on climate by directly reflecting the solar radiation to space and by modifying the shortwave reflective properties of clouds. Each year, increasing amounts of aerosol particles are released into the atmosphere due to biomass burning, dust storms, forest fires and volcanic activity. These particles significantly perturb the radiative balance on local, regional, and global scales. While the detection of aerosols over water is a well established procedure, the detection of aerosols over land is often difficult due to the poor contrast between the aerosols and the underlying terrain. In this study, we use textural measures in order to detect aerosols generated from biomass burning and dust storms.

The Operational Multiscale Environment model with Grid Adaptivity (OMEGA) is a new atmospheric simulation system that merges state-of-the-art computational fluid dynamics techniques with a comprehensive non-hydrostatic equation set. OMEGA is based upon an unstructured triangular prism grid that permits a horizontal grid resolution ranging from 100 km down to 1 km and a vertical resolution from a few tens of meters in the boundary layer to 1 km in the free troposphere. OMEGA also contains an embedded aerosol transport algorithm that permits the simulation at high resolution of the transport and diffusion of either grid-based aerosols or of Lagrangian parcels. OMEGA represents a significant advance in the field of weather prediction and aerosol transport. Current operational forecast models are scale- specific and have a limit to their resolution caused by their fixed rectangular grid structure. OMEGA, on the other hand, is naturally scale spanning and its unstructured grid permits the addition of grid elements at any point in space and time. This means that OMEGA can readily adapt its grid to fixed surface or terrain features, or dynamic features in the evolving weather. In addition, OMEGA can provide enhanced grid resolution in localized regions such as in the vicinity of a dust, smoke, or chemical cloud. The flexible grid adaptivity of OMEGA provides it with an important advantage over previous models. It permits the resolution of orographic and land/water boundary features improving the fine scale meteorological simulation and, in turn, the simulation of the aerosol transport. The coupling of a very high resolution (1 km) atmospheric simulation tool with an aerosol transport and diffusion model creates a flexible tool for a variety of applications and scales to be used anywhere that small scale features could have an impact on the local meteorology. Among these is aerosol transport in complex terrain and near land/water boundaries. In this paper, we will present an overview of the atmospheric simulation capabilities of OMEGA. We will discuss both its numerical techniques and its physics. We will also present an overview of the aerosol transport and diffusion model included in OMEGA; both its physical basis as well as its implementation on the adaptive unstructured grid that forms the basis of OMEGA. We will then discuss the application of this aerosol transport capability to the problem of simulating the transport of dust, smoke, or chemical clouds.

Satellite data retrieval algorithms almost always involve a large degree of model or simulation input. As an example, the satellite might provide a radiance or transmittance measurement that has to be unfolded to provide temperature or mass density. In order to convert transmittance into mass density, the operator must make some assumptions on the mass extinction coefficient and particle size distribution. These assumptions are often based upon climatological averages or upon simulation results. The Operational Multiscale Environment model with Grid Adaptivity (OMEGA) is a new atmospheric simulation system that merges state-of-the-art computational fluid dynamics techniques with a comprehensive non-hydrostatic equation set that includes both explicit and parameterized microphysics. OMEGA is based upon an unstructured triangular prism grid that permits a horizontal grid resolution ranging from 100 km down to 1 km and a vertical resolution from a few tens of meters in the boundary layer to 1 km in the free troposphere. OMEGA also contains an embedded aerosol transport algorithm that permits the simulation at high resolution of the transport and diffusion of either grid based aerosols or of Lagrangian parcels. OMEGA represents a significant advance in the field of weather prediction and aerosol transport. Current operational forecast models are scale- specific and have a limit to their resolution caused by their fixed rectangular grid structure. OMEGA, on the other hand, is naturally scale spanning and its unstructured grid permits the addition of grid elements at any point in space and time. This means that OMEGA can readily adapt its grid to fixed surface or terrain features, or dynamic features in the evolving weather. This feature also makes OMEGA a useful tool for satellite data retrieval and for the generation of synthetic satellite data. Synthetic satellite data is generated by recognizing that it is easier, in some ways, to simulate the performance of a sensor using the simulated environment and the sensor characteristics than to extract environmental information from the sensor data. This technique has been applied to generate simulated radar data as well as to produce a simulated photograph of an isolated cloud where the primary discrimination was the color contrast provided by the obscuration of the blue diffuse backscattered illumination by the cloud. The flexible grid adaptivity of OMEGA permits the accurate simulation of the satellite field of view thereby reducing the beam-filling problem that can cause major discrepancies in the data retrieval algorithm. Given the interaction of the model forecast and the data retrieval, the very high resolution forecasts possible with OMEGA also could improve existing retrieval algorithms. In this paper, we will present an overview of our concept of an analog sensor (or synthetic satellite data generation), and how the unique simulation capabilities of OMEGA factor into this concept.

Beginning in the early 1970's, the then Air Force Cambridge Research Laboratory initiated a program to develop computer-based atmospheric radiative transfer algorithms. The first attempts were translations of graphical procedures described in a 1970 report on The Optical Properties of the Atmosphere, based on empirical transmission functions and effective absorption coefficients derived primarily from controlled laboratory transmittance measurements. The fact that spectrally-averaged atmospheric transmittance (T) does not obey the Beer-Lambert Law (T equals exp(-(sigma) (DOT)(eta) ), where (sigma) is a species absorption cross section, independent of (eta) , the species column amount along the path) at any but the finest spectral resolution was already well known. Band models to describe this gross behavior were developed in the 1950's and 60's. Thus began LOWTRAN, the Low Resolution Transmittance Code, first released in 1972. This limited initial effort has how progressed to a set of codes and related algorithms (including line-of-sight spectral geometry, direct and scattered radiance and irradiance, non-local thermodynamic equilibrium, etc.) that contain thousands of coding lines, hundreds of subroutines, and improved accuracy, efficiency, and, ultimately, accessibility. This review will include LOWTRAN, HITRAN (atlas of high-resolution molecular spectroscopic data), FASCODE (Fast Atmospheric Signature Code), and MODTRAN (Moderate Resolution Transmittance Code), their permutations, validations, and applications, particularly as related to passive remote sensing and energy deposition.

The problems of optimum inversion in the presence of random noise are analyzed. Two main kinds of noise are considered: the random errors of measurements and random errors of physical model. It is studied the optimization of the numerical inverse problem solution concerning both noises. Using the statistical estimation ideas is discussed for this consideration. Specific features of every noise to influence on the limitation of information content of the optic experiment and on implementation of inversion are distinguished. The quantitative criteria to evaluate information content of input data and procedure of their interpretation are proposed. The latter is aimed to optimize the solution in presence of random errors of the model as well as errors of measurements and, moreover, to correct used model by the measurements being interpreted. An arbitrary accompanied and a priori information can be used. For example, a priori estimations of the sought and model parameters, correlations between them, non-negativity of values etc., can be included. The peculiarity of the inversion method is an essentially large number of variables and increased stability should be provided. The original iterative process of linear inversion characteristic to statistical optimizations are being proposed for this in the algorithm elaborating.

Vertically downwelling radiance spectra under clear sky were calculated in the 500 to 3000 wavenumber region. The calculations were carried out at wavenumber intervals varying between 20 and 80 cm^-1. The discrete ordinate radiative transfer code was used, along with the exponential sum fitting of transmission of water vapor, ozone, carbon dioxide, carbon monoxide, methane, and nitrous oxide. The spectral data base for the exponential sum fitting was obtained from the LOWTRAN7 transmittances. Observed vertical profiles of temperature, water vapor and ozone, obtained with conventional radiosondes and with Raman lidar, were used as inputs to the calculations. The sounding data were obtained during the Spectral Radiance Experiment (SPECTRE), held as part of FIRE Cirrus II in Coffeyville, Kansas, USA in November - December 1991. The calculated radiance spectra are compared to those calculated from LOWTRAN7. One spectrum is also compared to radiances observed near simultaneously during SPECTRE by the University of Wisconsin Atmospheric Emitted Radiance Interferometer instrument. The comparisons show a bias of 0.1 - 0.25 mW m^-2 sr^-1 (cm^-1)^-1 when averaged over the range of 560 to 3,000 cm^-1. This study was conducted in part in the framework of the Intercomparison of Radiation Codes in Climate Models.

During the `European Lidar Airborne Campaign' in 1990 the backscattered sunlight in the region of the O<sub>2</sub> A-band at 0.76 micrometers was measured above various clouds. The multispectral radiance measurements were performed with a nadir-looking device with a spectral resolution of (Delta) (lambda) equals 0.42 nm. A `Principal Component Analysis' was applied to about 140000 spectra, each consisting of 320 channels. It turns out that above clouds with optical depths (delta) _c > 10 or above clouds over oceans three spectral regions with (Delta) (lambda) approximately equals 2.5 - 5 nm contain the entire information of the high- resolved spectra. One of these intervals is located near the absorption band (window channel) while the remaining two cover the R- and the P-branch of the O₂ A-band, respectively. Based on this finding, the O₂ A-band channels of the planned `Medium Resolution Imaging Spectrometer' (MERIS, ESA), dedicated to the detection of the cloud-top height, are defined.

The simple semi-empirical expression, valid in the visible spectrum, for the diffuse irradiance calculations is presented. This relation is based on the experimental data collected in cloudless days of 1993 and 1994 over the southern Baltic Sea. In order to illustrate the accuracy of the proposed formula, the aerosol optical thickness is calculated. As a general conclusion it can be stated that the semi-empirical diffuse irradiance model can precisely estimate the spectral diffuse irradiance and the aerosol optical thickness.

Understanding the role of clouds in climate is one of the highest priority science objectives in the global climate change program. In particular, there has been a renewed interest in understanding cloud radiative interactions in the tropical regions. Although a number of studies have emphasized the importance of cloud optical properties on the earth's radiative energy balance, information concerning cloud optical depth and particle size is lacking. In this study, we use collocated AVHRR and ERBE measurements from the same satellite in order to understand the effect of cloud optical properties on the radiative balance of the earth- atmosphere system.

Clouds play an important role in forming the atmospheric radiation budget and are a specific factor for transformation of aerosol components. Cloud composition should be correctly enough taken into account to predict operatively the light propagation trough clouds. This matter is especially of high priority near the sources of industrial pollutions. Cloud microstructure would be the most variable quantity here, because of highly-disperse soot components ingress into clouds. Such components can influence greatly on spectral absorption and reflection characteristics of cloud cover and effect globally on climate. The consistent physical approach to predict the light propagation trough clouds requires the radiative model of clouds should include a complete set of the specific optical characteristics that, according to the radiative transfer equation, are sufficient to simulate the spectral absorption and reflection of cloud cover. Different approaches can be used to determine the above specific characteristics. The most promising one is appeared to be based on the retrieval of these quantities by airborne intercloud measurements of rather a narrow set of cloud light-scattering characteristics. The main object of this report is to investigate the method for solving an inverse problem of light scattering to design a radiative model of liquid-drops clouds comprising soot components. The investigation is carried out within the scope of mathematical simulation.

The atmospheric abundance of several radiatively and chemically active trace gases is changing. In order to understand the proper atmospheric processes as transport of trace gases and chemical reactions it is necessary to perform simultaneous and global observations of these trace constituents. This can be achieved with Fourier Transform Spectrometers (FTS) detecting wide spectral i ntervals i n the mid-Infrared (IR). During the last ten years several FTIR experiments for measuring trace species have already been carried out by different scientific groups (overview see Fischer 1992). In the following only more recent measurements of FTIR-instruments will be treated . There exist u ncooled mid-IR experiments wh ich yielded still very i nteresti ng scientific results even if they provide only sun occultation measurements, i.e. these measurements are restricted to a limited number of occasions during dawn and dusk. A widely known FTIR experiment of such type is the so-called ATMOS (Atmospheric Trace Molecule Spectroscopy) Project. During several Spacelab flights a large number of sun occultation measurements have been recorded and to a far degree already analysed. The results are concentration profiles of about 30 trace gases in both the Northern and Southern hemisphere. Cooled FTIR instruments have already been operated aboard stratospheric balloon gondolas by some scientific groups and aboard a Transall aircraft. The recorded emission spectra are of high quality and allow the determination of vertical profiles/column amounts of many trace species, e.g. of the main reactive and reservoir nitrogen compounds. For example, the MIPAS (Michelson Interferometer for Passive Atmospheric Sounding) experiments provided remarkable results with respect to ozone research in the Arctic, e.g. the spatial and time-dependent distribution of the important reservoir species CIONO2. The talk will give an overview on FuR instruments in operation, a selection of the more recent measurements and the corresponding scientific results. Keywords: Remote sensing, trace species, Fourier Transform Spectrometers, midInfrared

The unified retrieval approach is applied to Defense Meteorological Satellite Program (DMSP) 5D-2 microwave sensor data (i.e. SSM/I, SSM/T, SSM/T-2) combined with that from visible and infrared cloud imager data in a physically based approach to optimally infer the state of the atmosphere/ocean background. Surface emissivity and cloud liquid water are treated explicitly and are retrievables. The system is compatible with numerical weather prediction data assimilation and can utilize forecast first guesses in the fully interactive mode. We present validated retrievals of atmospheric temperature profiles, water vapor profiles, cloud liquid water, surface temperature and surface emissivity obtained from DMSP sensor data from the F-11 spacecraft for which limited radiosonde collocations are available. We note the utility of the retrieval system, specifically the residuals, to diagnose: phenomenon omitted from the forward problem such as cirrus cloud, spatial inhomogeneities of retrieved quantities such as water vapor and cloud, and data inconsistencies such as biases between channels and instruments. The retrieval system developed is equally applicable to other meteorological sensor payload combinations such as the OLS-SSM/IS, AMSU/MHS, and AVHRR/HIRS.

Clouds exert a strong influence on the distribution of heating and cooling at the Earth's surface and within the atmosphere. This has long been recognized in climate studies and in the development of General Circulation Models'2 (0CM), much less in Numerical Weather Prediction (NWP), until the extension of useful forecast length and the diversification in the products range has given new impulse to the development of more accurate parametrizations of diabatic processes3. When the new ECMWF prognostic cloud scheme becomes operational, the ECMWF model will acquire some capability to assimilate cloud-related information. To prepare further development, calibrated and earth located, or raw, Tiros-N Operational Yertical Sounder (TOYS) radiances are used as a diagnostic tool to evaluate the ability of sequences of short range forecasts, from different models, to simulate some of the features in the measured data. The present exercise is therefore quite different from the diagnostics which employ long term integrations of a model to test its average properties against some independent data set. The main difference between the raw TOYS radiances used in the present exercise and the radiances used operationally at ECMWF is the cloud clearing process, whose aim is to identify and eliminate the data affected by clouds. Other differences will be discussed in section 2. Since the signature of clouds on upwelling atmospheric radiance is quite marked at visible and infrared wavelengths, inadequacies in model representations of the three dimensional structure of cloud cover and cloud liquid water are easily identified. One can therefore make a detailed examination of aspects of the hydrological cycle and energetics in current NWP models and in GCMs. Although the importance of clouds in NWP lies mainly in their radiative effects, it is common practice to compare model cloud properties, such as cloud amount (either total or for thick layers) with "equivalent" quantities retrieved from data sets which are independent of the model, and are frequently derived from satellite radiance data68. In such a comparison the same name (for example total cloud amount) may be given to quite distinct quantities, obtained using different underlying assumptions9. Great care is needed when comparing these satellite products among themselves and with model products'°. On the other hand when simulated radiances, obtained using all pertinent model information, are compared to measurements, one needs to be aware of the assumptions used in the simulations themselves. The evaluation of the errors is in both cases a delicate procedure, but the error estimation is particularly difficult for retrieved products such as total cloud cover. In the following the results of a comparison between a selected set of raw TOYS radiances and simulations using the required model output fields is presented.

In recent years the concern about the impact of the most important greenhouse gas, atmospheric water vapor, on the earth's climate was a driving force for many scientific projects. As the attention on atmospheric water vapor increased, the need for global, accurate and consistent data sets over long periods increased simultaneously. Satellites provide an ideal platform for such observations. For the geostationary satellite Meteosat the requirements of global and consistent measurements are certainly fulfilled, though the accuracy could be subject to improvement. When working with satellite data quantitatively an accurate calibration of the data is essential. Since a direct, on board calibration of Meteosat satellite imagery is not possible, a vicarious calibration is necessary. The calibration of the Meteosat water vapor channel is performed using information from radiosonde ascents. Changes in the operational calibration method implemented on the 4^th February 1994 yield a more accurate and stable calibration. The water vapor channel is used operationally to retrieve humidity information within the upper troposphere. Within the relevant layer (600 - 300 hPa) a mean relative humidity is derived. Such product may be used in the initialization of numerical forecast models, in the validation of short- and long-term models and in eliminate monitoring of the upper-tropospheric humidity. Monthly averages of the Upper Tropospheric Humidity (UTH) fields reveal a relationship to the large-scale global circulation. Four times a day cloud motion vectors are determined from Meteosat water vapor imagery. The individual fields show the local wind distribution at a given time. Yet, the monthly mean of such wind fields reveal distributions on larger time and space scales. The divergence field from such a monthly mean wind field shows a correlation with the mean UTH field indicating that the UTH is largely determined by the large-scale atmospheric dynamics.

The hydroxyl radical (OH) is a principal oxidant in the earth's atmosphere. The OH radical plays an important role in the destruction cycles of ozone and it reacts with greenhouse gases like methane and chlorofluorcarbons. To increase our knowledge about these chemical reactions in the atmosphere there is a need for accurate determinations of the OH concentration profile. For this purpose a Fabry Perot (FP) interferometer can be used on board high flying aircraft and, with modifications on satellites. The FP is optimized to measure the emission of OH in the far infrared, at 118.45 cm^-1. The FP can also measure the emission of other species, like H₂O and O₃. Our computer simulations show that detection is possible when observations are performed above the tropopause, where water vapor concentrations are low. Spectra will be presented which show the expected OH signal at 13 and 20 km observation altitude with a resolution of 0.015 cm^-1, a noise level of 10^-13 - 10^-15 Watt Hz^-1/2 and a sounding elevation of 10 degrees.

An excellent instrument to observe the emission lines of OH in the far infrared is a cooled scanning Fabry-Perot (FP) where the orders are selected by means of a grating monochromator. A compact liquid helium cooled prototype has been built for airplane missions. A similar device is the OHIO instrument, proposed for limb observations from a satellite and applying detectors in conjunction with space-qualified refrigerators. Compared with heterodyne receivers, an FP at 80 K offers a comparable sensitivity in the 100 (mu) wavelength region and much better values with bolometric detectors optimized for lower temperatures. Compared to satellite-borne Fourier Transform Spectrometers (FTS), the FP instrument is more sensitive at 80 K; the FTS measures many superfluous spectral elements. The FP is superior at temperatures low enough to allow photon-noise-limited detector operation.

In the interferometers for atmosphere spectrometry from the space, special scanning mechanisms are required. In the following an original solution for this device, based on elastic suspensions and position servo-control is described. Some experimental results, concerning the static accuracy, obtained on a technological model of the mechanism, are reported, together with simulations of its dynamic behavior.

Since December, 1992 a continuing data bank of measurements of global, direct and diffuse solar irradiance for clear skies is being compiled at Valencia (Spain). The measurements are performed with the commercial LICOR spectroradiometer from 300 - 1100 nm with a spectral resolution of 6 nm. As a complementary work to irradiance measurements an extensive comparative task is being carried out between these experimental data and modelled ones. The chosen model is a detailed narrow band model developed at the Laboratoire d'Optique Atmospherique of the Universite of Lille (France) which uses a two-flux method to solve the radiative transfer equation. The multilayer model (17 atmospheric levels) considers the atmosphere as a plane-parallel absorbing and scattering medium where total, direct and diffuse spectral irradiance values are calculated at each level. Preliminary analysis shows promising results but the difficulty in modelling aerosol properties to fit experimental data is clearly manifested.

The Mount Pinatubo eruption (15 degree(s)07'N, 120 degree(s)20'E) in the Philippines of June 1991 was among the largest volcanic eruptions of this century in terms of effects on stratospheric aerosols. The activity culminated in a paroxysmal eruption on June 15th and developed a giant umbrella cloud which introduced a large amount of ashes and gases into the stratosphere. The high frequency of coverage of the NOAA (USA) and GMS (Japan) weather satellite enables a global monitoring of the rise and spreading dynamics of the Pinatubo volcanic cloud into the atmosphere. By integrating the maximum eruption height and the spreading rate over time, the total volume of pyroclastic material has been estimated to range between 3 and 4 Km³. Image processing techniques such as difference T4-T5 and Principal Component Analysis have been applied the discrimination between volcanic cloud, ice cloud (cirrus) and clouds containing water vapor and water droplets. These detection techniques provide an operational tool for tracking the horizontal dispersion. The results can be used in the fields of monitoring long distance transport of the volcanic cloud over land and sea, aircraft safety and global atmospherical impact.

A database application system was developed to provide aerosol parameters for radiative transfer calculations. The system is based on the aerosol model of d'Almeida (1991) and is directed in its first step to maritime applications. The user can choose location and time for his radiative transfer calculations. The system offers proposals as tables and curves for optical aerosol parameters and aerosol height distributions. Depending on the user requirements the data (e.g. mixing ratio, height distribution) can be modified.

The method and computation algorithm have been developed to assess the effect of scattered light on a signal received by a sun photometer while a cloud optical thickness is measured. The approach provides estimations of impact of scattered light for any receiver field of view with allowance for angular dispersion of sunlight. The accuracy of measurements has been investigated as a function of zenith angle of the Sun, receiver field of view, wavelength, cloud optical thickness. The effect of aerosol scattering is an undercloud layer on the accuracy of cloud optical thickness measurement has also been considered.

Contrails are produced by airtraffic and thereby form an anthropogenic component of cloudiness. They typically appear in height levels between approximately 6km and 12km. Their temperature is typically between -40°C and -60°C. Due to these low temperatures their greenhouse effect is potentially high and the question arises to which extent they influence the climate (U. Schumann et a!., 1 994). A prerequisite for an estimation of this climatic influence is a knowledge of the cloud cover which is due to contrails. Our study presents two methods of determining contrail cloud cover using NOAA-AVHRR satellite data. The temperature difference image between channels 4 and 5 at 1 1 .tm and I 2tm enhances the appearence of high level thin clouds and is therefore used as a basis of the analysis. The first method is 'manually', the second by application of an automated detection. Both methods are applied on a scene over Central Europe and compared to each other. Chapter 2 describes the manual analysis and its results, chapter 3 the automated method and its limitations which exist up to now and chapter 4 shows the comparison of the two methods.

Light scattering by macroscopically volume element, which consists of arbitrarily shaped particles with arbitrary square integrable probability density function over orientations is considered. The T-matrix approach and the quantum theory of angular momentum are used to develop a rigorous analytical method to compute the elements of the Mueller matrix, the Stokes vector of scattered radiation for arbitrary number of incoherent sources of the incident radiation, and scattered flux into any conical solid angle. Some consequences of the analytical results are presented.

The anomalous diffraction approximation is used to prove optical equivalence of randomly oriented ellipsoidal, polydisperse randomly oriented spheroidal and polydisperse spherical particles for extinction, scattering and absorption cross sections. The distribution functions over the sizes, and shapes of particles are presented. The correctness of proved statements is illustrated by numerical results with help EBCM (extended boundary condition method) for spheroidal particles and Mie theory for spherical particles.

This paper presents comparisons between aerosols measurements and aerosols in Lowtran7. Desert aerosols are natural aerosols that can be found throughout the atmosphere; they are a mixture of different kinds of materials. The model used in Lowtran7 separates the desert aerosol into its three major components: carbonaceous particles, pure quartz and quartz contaminated with a small amount of hematite. Recent measurements have been carried out for three samples of desert aerosols from the Middle East and give prominence to a complex composition of the sands. Thanks to an accurate analysis, we present features of the sands and compare them with values introduced in Lowtran7 desert aerosol model.

The Air Force has long maintained an `exact' accelerated line-by-line radiative transfer model, the Fast Atmospheric Signature CODE (FASCODE), appropriate for applications in both the laboratory and any arbitrary line-of-sight in the atmosphere. The current version of FASCODE, FASCOD3, is fully compatible with the HITRAN92 database, including access to the temperature-dependent cross sections for heavy molecules (e.g. CFC's etc.). Some new features of FASCOD3 are: line coupling algorithms for both 15 micron CO<sub>2</sub> and the mm lines of O<sub>2</sub>; non-local thermodynamic equilibrium models; updated H<sub>2</sub>O continuum; multiple scattering capability; and laser options for lidar modeling applications. FASCOD3 is increasingly being used as a high resolution remote sensing data analysis tool from microwave and infrared (IR) to ultraviolet (UV) spectral ranges. The Moderate Resolution Atmospheric Radiance and Transmittance Model (MODTRAN) is a `first principle' band model with a nominal spectral resolution of 2.0 cm^-1. Model parameters are derived directly from the HITRAN database. Standard 2-parameter Curtis-Godson approximations are used for H₂O, CO₂, etc., and 3-parameter Goody approximation is used for O₃. The current version of MODTRAN, MODTRAN3, encompasses all the capabilities of LOWTRAN, and contains many important elements that many other band models do not incorporate, including: Voigt line shape; spherical geometry; solar and lunar source functions (irradiance); internal aerosol, clouds, and rain models; single and multiple scattering; default atmospheric profiles. Because of its speed advantage over FASCODE, about a factor of 100, and ease of use, MODTRAN3 has been and will continue to be an effective tool for atmospheric spectral heating/cooling rate calculations and atmospheric corrections in earth surface sensing and imaging. Because of the large user based of FASCODE and MODTRAN, it is very important to continue to improve and validate those codes. This paper presents the validation of FASCOD3 and MODTRAN3 in the context of SPECTRE and ITRA. Important considerations (such as water vapor continuum, frequency-dependent sea surface emissivity in the IR window region, and spectral resolution of MODTRAN3) in the comparison of model calculations with high resolution interferometer measurements will be discussed.

Balloon borne observations of polar stratospheric cloud (PSC) were made in Kiruna on February 4, 1990 during CHEOPS 3 campaign. The measurements of radiance and polarization of the scattered sunlight in the horizontal plane, for the whole range of scattering angles in two polarimeter channels (850 and 1650 nm) were performed at different altitudes of the balloon. The phase function behavior indicated that the top of the PSC (at the altitude about 24 km) was composed of rather large, micron sized particles but difficulties in retrieving the measurements, especially the polarization ratios, from calculations for spherical particles suggested that they had irregular shapes. The purpose of this study is to retrieve the scattering characteristics of the top of PSC assuming the nonspherical particle shape. The calculations have been performed using the discrete dipole approximation. From among the shapes studied the best agreement has been obtained for rather elongated spheroids.

Temperature and humidity sounding of the atmosphere from polar orbit meteorological satellites is currently carried out by combining data from infrared and microwave sounding radiometers. The High Resolution Infrared Sounder, the current operational infrared sounder, has 20 channels and accomplishes spectral separation with a filter wheel. This small number of channels limits the accuracy and vertical resolution of sounding and is thereby one of the limiting factors to the accuracy of numerical weather prediction. Studies have shown that infrared sounders with 1000 to 2000 channels can provide the desired 1 K accuracy and 1 km vertical resolution. NASA is developing the Atmospheric Infrared Sounder, which uses gratings for spectral separation. There are several infrared sounders under development that use Michelson interferometers for spectral separation.