Recent progress in the understanding of imaging when an object is both illuminated and viewed through the same random screen is reviewed, with particular emphasis on the use of non-redundant apertures.

Simultaneous measurements of the spatial intensity covariance of laser scintillation and the variance of integrated intensity collected by multiple circular apertures provide sufficient information to determine the form of the atmospheric turbulence spectrum over short spatial and temporal scales. This ensures local stationary of the turbulence. The estimated parameters of the turbulence spectrum (level of turbulence C2n and inner scale) over these stationary events provide accurate estimates of the intermittency of atmospheric turbulence. The advantages of laser scintillation measurements include spatial averaging that improves statistical accuracy, true measurements of spatial statistics without the need to convert temporal statistics to spatial using Taylor's hypothesis, remote sensing of atmospheric turbulence which reduces contamination by support structures, and rugged instrumentation for reliable measurements. New uses of laser scintillation measurements will be discussed.

New spectra models have recently been developed for the spatial power spectra of temperature and refractive index fluctuations in the atmospheric boundary layer showing the characteristic 'bump' just prior to the dissipation ranges. Theoretical work involving these new models has led to new expressions for the phase structure function associated with a plane optical wave, although most experimental work has involved spherical waves. Following techniques similar to those used for the plane wave analysis, new expressions valid in geometrical and diffraction regimes are developed here for the phase structure function of a spherical optical wave propagating through clear-air atmospheric turbulence. Useful asymptotic formulas for small separation distances and the inertial subrange are derived from these general expressions.

Expressions for two-frequency correlation characteristics have been constructed using multiscale solutions for high-frequency random propagators. Analysis of the spectral characteristics of plane and spherical waves was performed.

Although optical turbulence is usually modeled with micrometeorology, it is shown here that this can be done successfully with macrometeorology using meteorological parameters measured with standard weather stations and predicted in standard weather forecasts. This makes it possible to predict C2n according to the weather forecast. Two experimentally-derived models are developed--one for practical use and the other for scientific understanding. Correlation of prediction with measurement is 90% or more, over large dynamic ranges of meteorological parameters. One interesting aspect of these measurements is the statistical evidence that scintillations are affected by aerosols, particularly under conditions of high total aerosol cross sectional area. This may be attributed to increased refractive index changes encountered by radiation which penetrates through the aerosols. In addition, validity of the models was examined, and experimental comparisons in two very different climates and surface conditions are presented. High correlation is found in both cases between prediction and measurement.

Several sets of experimental micrometeorological data were used to obtain estimates of vertical profiles of C2n for damp unstable conditions. These data consisted of two types: (1) latent and sensible heat fluxes and (2) vertical profiles of wind, temperature, and specific humidity. Estimates of the scaling lengths of virtual potential temperature and specific humidity obtained from these data were used to calculate their vertical gradients and, in turn, to estimate C2n in accordance with Tatarski. Results from these data sets are presented to emphasize the relative contributions of temperature and moisture to C2n.

Wavefront curvature sensing and compensation may lead to lower cost adaptive optics systems optimized for low order correction.

The iterative expectation-maximization (EM) algorithm for identifying unknown blur, noise, and image parameters under Gaussian modeling assumptions is described. A new form of the equations updating parameter estimates is given, from which convergence conditions and symmetry properties of the parameter estimates are derived. The frequency domain resolution defined by the digital image is not appropriate for accurate parameter estimation. Instead, a version of the EM algorithm with frequency resolution appropriate for the blur point spread function (PSF) is proposed. Results are presented from a test of the reduced resolution algorithm, in which the importance of the initial PSF is studied.

The interaction of exposure time and system noise with angle-of-arrival measurements taken with the Atmospheric Turbulence Measurement and Observation System (ATMOS) is presented. Of particular interest is the effect at short exposure times when viewing stellar images. Both measurements and simulated results at various exposure times are shown. An analysis is discussed of the effect on the variance of angle-of-arrival data for short exposures when the frame rate produces 'dead' time between samples. The dependency of the exposure time/system noise interaction on deriving accurate transverse coherence length (ro) and power spectral densities (PSD) of angle-of-arrival from centroid measurements using the ATMOS is presented.

A full 4-D wave optics computer code, GRAND, has been developed to study the atmospheric propagation of a high energy laser beam. We perform detailed calculations of non-linear interactions of turbulence and thermal blooming and their effects on high energy laser beam propagation. The simulation code includes sophisticated models of realistic adaptive optics compensation systems that are used to correct the atmospheric distortions. The propagation code has been used to investigate beam quality amelioration by natural atmospheric dynamics. Arbitrary engagement geometry can be included in code simulations. We present an overview of the propagation code and discuss its architecture, operational features and some recent applications. The 4-D wave optics propagation code is available for collaborative studies in atmospheric optics.

A new technique for sensing refractive-turbulence profiles is described. The technique is based on performing a spatial correlation of the measured wave front phase from two reference sources. This technique is unique in that the correlation properties of the wave front phase are used instead of the more common approach of using optical scintillations. The geometry between the reference sources and the two wave front sensor apertures is arranged such that the two optical paths cross at some point in front of the sensor plane. The wave front phase for each reference source is reconstructed from the measured wave front sensor data. A spatial correlation of the two reconstructed phase maps is performed. From this correlation, a measure of the structure constant of the refractive index fluctuations C2n can be extracted. The resolution of the technique depends on the angle between the optical paths, the spatial frequency response the wave front sensors, and the size of the wave front sensor apertures. For sensing the vertical profile of C2n, resolution on the order of 50 meters can be obtained by using sources separated by 0.17 degrees.

Various sources of noise associated with angle of arrival measurements are discussed and their effect is evaluated and compared to experimental determinations of system noise. The predominant source of system noise is found to arise from the charge transfer process on the CCD detector itself. Image motion variance measurements are linear to within 3% over the range of image motion encountered in ro measurements with a correlation coefficient greater than 99% when compared to a known reference.

Two different methods (models) are used for simulating vertical profiles of extinction and backscatter coefficients in very low stratus clouds and subcloud regions. The first method makes use of a microphysics model developed by the authors. This model simulates drop size distributions as functions of height above the ground while considering the limitation on drop growth rate by the rate of diffusion of water vapor. Mie theory is then used with the drop size distributions and complex indices of refraction of water and material from condensation nuclei to simulate vertical profiles of extinction and backscatter coefficients. Comparison is made of profiles calculated by assuming the drops to be pure water with profiles calculated by considering the drops to be a combination of water and dry material. The second method uses vertical profiles of relative humidity and temperature simulated by the first method. These profiles are used to modify the reference level drop size distribution to simulate changes with height assuming the drops are in equilibrium with their immediate environment. The restriction of diffusion rate is not used in the second method. Applying Mie theory to the resulting drop size distributions yields a second set of profiles of extinction and backscatter coefficients. Profiles of these coefficients simulated by the first method appear more realistic than those simulated by the second method. Relationship between backscatter and extinction coefficients simulated by the first method also is shown.

Models for aerosol in the marine atmospheric boundary layer are reviewed with emphasis on applications to optical and IR propagation. This includes aerosol extinction and effects on the refractivity through coupling between aerosol and temperature and humidity profiles. Atmospheric processes and influencing the aerosol concentrations are discussed. Both empirical models and physical models are considered. In addition, original material is presented on modeling extinction profiles in the marine atmospheric surface layer.

The Navy Oceanic Vertical Aerosol Model (NOVAM) was evaluated by making nearly simultaneous measurements of atmospheric structure with an airborne lidar and Particle Measuring Systems, Inc.'s aerosol spectrometers. Profiles of measured aerosol-size distributions were scaled to the lidar returns, and the calculated extinction coefficients were compared with the NOVAM predictions. NOVAM-predicted extinction coefficient profiles have similar structure to the aircraft-measured profiles but underestimate the absolute value. When scaled to visibility, good agreement exists between the measured and predicted profiles. Agreement between the predicted and measured extinction coefficients is strongly dependent on air-mass factor.

The subject of this study is to validate the Lowtran 7 code, particularly its Navy Maritime model, in a mediterranean marine atmosphere. Measurements of aerosol profiles were made on board of a ship during a Mediterranean cruise. Their behavior was analysed under different meteorological conditions and measurement locations. On the other hand, aerosol, meteorological, and 3-5 micrometers broadband transmission measurements were performed in the Toulon's roads. The devices used for this experimentation were a Knollenberg's spectrometer for aerosol profiles and an IR broadband transmissometer along a path of 8 Km above the sea surface (50 meters high). The results lead to the conclusion that the Navy Maritime model is unadapted to a mediterranean atmosphere, probably due to the proximity of the coasts. So, the use of Lowtran 7 code in this area needs a more efficient model.

This paper gives a general formalism which can be employed to study the intensity of a pulse propagation through burning particles. To correct for multiple scattering effects, a modified extinction cross section obtained from multiple scattering theory can be used in the equation of radiative transfer to compute the specific intensity. Results of the computed reduced intensities using the multiple scattering efficiency and the extinction efficiency of a single particle are given.

An effort to quantify effects of weather on coarse aerosol concentration and total cross- sectional area per unit volume, so as to permit their prediction according to weather forecast, has begun. Correlations of prediction with measurement are on the order of 92% and 88% respectively. Relative humidity is the dominant parameter. Based on this work, further development is planned so as to predict overall size distribution, scattering coefficients, and aerosol MTF, according to weather forecast.

Thermal image quality depends upon properties of hardware, atmosphere, and thermal contrast in target plane. Weather affects both MTF of the atmosphere and thermal contrast in target space. Quantitative relationships have been determined relating overall thermal image quality to weather for imaging of passive targets and are suggested as a criterion for forecasting relative quality of thermal imagery according to weather forecast.

A unique research program is being conducted for passive, remote sensing of horizontal contrast transmittance and the modulation transfer function (MTFA) of desert atmospheres, including the DC, low, and high spatial frequency MTFA components that are attributed to contrast, aerosols, and turbulence, respectively. The research makes it possible, for the first time, to measure the overall low spatial frequency response of the aerosol MTFA components of actual desert atmospheres. The required measurements are based on utilizing digital image processing of remote video scenes which include two optically-identical castellated targets, contrasted against the horizon sky background. Ratios of apparent contrast and MTF measurements of these two targets are used to determine the contrast transmittance and MTFA in the atmospheric region between the two targets, independent of the imaging system and the actual optical properties of the targets. The experimental technique is described along with current MTFA measurements.

The atmospheric medium consists of aerosols and gases that can greatly affect the propagation of electromagnetic energy. Because these constituents, as well as temperature and turbulence, can vary greatly in time and space, propagation codes based on standard models of atmospheric parameters, climatological records, or limited measured data may not adequately simulate the propagation environment exposed to EO/IR sensors and systems. Lidar techniques are reviewed and data examples are presented that illustrate lidar capabilities for measurement of atmospheric transmissions over extended paths. The lidar also provides measurements of atmospheric structure and quantitative data on aerosol and gas concentration and temperature distributions needed to evaluate the propagation environment.

The use of backscatter lidars as a research tool to remotely sense atmospheric parameters has been well established. But, the use of lidars as an operational tool to predict the performance of electro-optical (EO) systems during periods of adverse weather has not. A model correlating lidar derived atmospheric transmission and FLIR (forward looking infrared) performance has been developed and an international field program, FLAPIR, has been conducted to collect the data necessary to evaluate the model.

As a beam of light propagates through the atmosphere, scattering by aerosol particles causes the beam profile to broaden. A multi-field-of-view (MFOV) lidar has been developed which makes simultaneous measurements of the energy directly backscattered from the central beam and multiscattered signals arising from the broadened part of the beam. The direct backscatter signal constitutes a conventional Mie lidar signal. Measurements made along a near horizontal path in haze, fog, and rain are presented. The results show that the multiscattered signals are strongly influenced by the extinction coefficient and the size of the aerosols. Thus the multiscattered signals, together with the direct backscatter signal, contain more information about the aerosols than is available from a conventional single field of view lidar.

A wide spectral range transmissometer has been developed at the Phillips Laboratory for long path atmospheric measurements. It was designed to provide narrow band spectral coverage in three bands between 0.4 and 14 micrometers by programmed scanning with circular variable filters. The system uses three detectors so that three spectral regions can be sampled simultaneously. Two sources, a black body and a ribbon filament lamp, are combined in a single beam. A two-way path is provided by use of cats-eye retro-reflectors. Transmitter and receiver are combined in the same unit and, by using multiple retro-reflectors located at different positions, the path length can be readily changed by realignment of the unit. This increases the dynamic range of the system and makes possible measurement of rapidly changing atmospheric conditions. The transmissometer was recently used to provide transmission measurements as part of the FLAPIR (forward-looking infrared and lidar atmospheric propagation in the infrared) field measurement program at Brunswick Naval Air Station, which had the purpose of determining the feasibility of using lidars to predict the effectiveness of infrared imaging devices. Measurements were made primarily in fog and rain conditions.

Optical communications systems between earth and planetary spacecraft are being developed for use in the next century. Ground-based receivers must be prepared to contend with a variety of atmospheric conditions. Improved weather models will enable more reliable predictions of system performance. The atmospheric visibility monitoring (AVM) project has been designed to enhance present models and produce joint visibility statistics for multiple sites in the southwestern United States. Three autonomous observatories will be deployed to measure atmospheric conditions based on observed starlight. A preliminary model predicts that from three sites in the southwestern United States with a low correlation of weather patterns, at least one site should have clear skies adequate for optical communications 94% of the time. Data from the observatories will give clear-sky and transmission statistics for three particular sites chosen for a high probability of clear skies. Transmission measurements will be taken using broadband astronomical filters as well as narrowband filters corresponding to laser wavelengths being considered for communications. Ground-based data will be compared to satellite imagery to determine the correlation between satellite data and ground-based observations.

This study provides a method for quickly calculating the slant and horizontal two-way atmospheric transmittance of a carbon dioxide laser radar operating at 10.591 micrometers . Calculations using this method are performed comparing the two-way transmittance for five atmospheric models: tropical; mid-latitude summer; mid-latitude winter; subarctic summer; and subarctic winter. The two-way transmittance is calculated using an exponential expression for the scattering and absorption attenuation coefficients of the atmospheric molecules and aerosols. The expression, obtained by fitting existing atmospheric attenuation coefficient data, are integrated over the slant range of interest. Results show that the attenuation decreases when the altitude of the laser radar increases for horizontal- and upward-looking line of sight. Results also show that the attenuation increases with increasing altitude and slant range when the laser radar is looking downward toward the earth. The method described can be applied to other available attenuation coefficient data at different laser frequencies.

Accurate models are needed to represent both the local lines and the continuum absorption in spectral ranges of interest. Additionally, accurate experimental data are needed, under different conditions of pressure and temperature, to test the validity of various models. Experimental data are obtained from a BOMEM fourier transform spectrometer (FTS) with a high pressure-high temperature cell and a 10-m white cell. Absorption coefficients are determined for gas mixtures (H2O, CO2, N2, O2) for pressure up to 60 atm and temperatures up to 600 K. At high pressure, the Lorentzian approach fails, and semi- empirical models are used to represent local line and far wing phenomena. The far wing nature of the line shape theory of Birnbaum is used to represent the water vapor continuum. Comparisons are made between our experimental data and synthetic spectra based on the HITRAN data base and Birnbaum's line shape for several atmospheric transmission window regions. Implications concerning atmospheric propagation are emphasized.

A forward scatter meter is a convenient and accurate method to measure the local visible extinction coefficient in fog and haze. The device works by illuminating with near-IR light a sample volume of air and measuring the intensity scattered in the angular range of 27-42 degrees. The scattered intensity is well correlated to the extinction coefficient of fog regardless of the fog droplet size distribution. The forward scatter meter also gives meaningful measurements during rain and snowfall. A comparison of extinction measurements made with a narrow beam transmissometer and the forward scatter meter have been made. The results show that during rain, the forward scatter meter extinction coefficient is from 25-50% greater than that measured with the transmissometer. During snowfall the forward scatter meter extinction coefficient is about 10-40% less than that measured with the transmissometer. These results can be used to define correction factors so that the forward scatter meter can be used to estimate extinction during rain and snow as well as during fog.

An approach to a two-dimensional vector thermal radiative transfer equation for a layer of spatially inhomogeneous atmospheric rain and cloud is developed. By using the Fourier transformation in discrete form and matrix expression, 1-D radiative transfer equation is derived. Employing an iterative approach, the zeroth- and first-order solutions from integral radiative transfer equation are obtained. Numerical results of polarized brightness temperature of atmospheric rain and cloud with a Gaussian fractional volume are obtained. The functional dependence of brightness temperature to inhomogeneous scale, frequency, polarization, and other parameters are discussed.

The limits of image quality through the atmosphere depend on the overall atmospheric modulation transfer function (MTF) cutoffs. The first spatial frequency cutoff, or 'knee,' of the overall atmospheric MTF curve depends on aerosol MTF. The second spatial frequency cutoff, limited by threshold contrast required in the output image, also depends on turbulence. Measurements of atmospheric MTF over a 5.5 km horizontal path near the ground were made for a large range of spatial frequencies at several wavelengths in the visible and near-IR spectrum along with measurements of turbulence by a passive edge wander technique. On-line measurements of particulate size distribution were made using a Particle Measurement System, Inc. (PMS) probe, and on-line meteorological data (air temperature, relative humidity, wind speed, wind direction, and solar flux) were obtained from a weather station a few meters away from the imager. The experimentally-derived MTF curve is compared to well-known models for turbulence MTF and aerosol MTF. The determination of aerosol MTF from macroscale parameters is crucial to prediction of image quality through the atmosphere and can be implemented in image restoration.

The main aerosol effect on imaging performance is brightness reduction through scattering losses. This is fairly well understood and modeled. However, one phenomenon that is almost always overlooked is the blurring of images that can result from forward-scattered radiation, i.e., the radiation reaching the image plane after being scattered by airborne particles. This paper describes a simple experiment to measure the visible point-spread function, or the image of a point source, through fog and rain. The images were recorded with a CCD camera. Frame addition was used to reduce the statistical noise. Series of images were made with different neutral density filters and later recombined to increase the dynamic range beyond the 8-bit gray-level range of the frame grabber. The results show the effects of range, particle density, and particle size. The measurements are generally in good agreement with model predictions. As it turns out, the aerosol blurring effects are important only for rain and for some advection fogs with a sufficient number of particles in the size range of $OM 100 micrometers.

Analytical models of the spatial power spectral of temperature fluctuations and refractive index fluctuations that feature a 'bump' at high wave numbers have recently been developed. These models, based on the effects of intermittency, evolved through an approximation for the intermittency effect that led to a two-term modification of the standard Tatarski spectra models. This paper examines the intermittency correction with experimental data and standard spectra models. Specifically, more accurate approximations to the intermittency effect that lead to a four-term modification of the Tatarski model are developed. A comparison between these two modifications of the standard Tatarski spectrum is discussed in the context of theoretical studies concerning optical propagation through atmospheric turbulence.

The effect of anisoplanatism in adaptive optics is considered. Phase Derivative Adaptive Optics (PDAO) is suggested to reduce anisoplanatism and to increase the effective region of compensation. The results show that PDAO is a prospective method to solve the beacon's problem in adaptive optics.

This paper discusses propagation of paired field measures and their spectral transforms in a random medium. Using a directional source initial condition, a computable algorithm for the intensity correlation characteristics of finite aperture radiation is presented.

Low energy laser propagation through moderate turbulence is considered. Propagation of lasers through the surface boundary layer at ranges greater than a kilometer for visible wavelengths often involves intermediate level turbulence effects. Although propagation through weak turbulence is described through a series of integral equations and propagation through strong turbulence is understood in the asymptotic theory, propagation through intermediate level turbulence is usually described through heuristic theories. In this paper, techniques are described to simulate beam spread, wander, scintillation saturation, aperture- averaging, and an aperture-averaged scintillation distribution. Inner-scale effects are discussed throughout. Hill and Clifford theory is parameterized and adjusted to obtain an irradiance variance saturation curve that compares well with measured data of Coles and Frehlich, and Churnside. An empirical aperture-averaging curve compared well with intermediate turbulence data collected by Churnside.

Correlation of bifrequency beam propagating in a folded turbulent atmospheric path reflected by plane mirror is analyzed and a theoretical expression of correlation coefficient is developed by using Kolmogorov turbulence model in this paper. It is noted that in this optical path the greater the difference between two frequencies is, the worse the correlation will be. The correlation coefficient for folded path is larger than that for direct path when the Fresnel parameter becomes smaller than a certain value and vice versa. The aperture averaging effect has the same characteristics.

Deconvolution from wavefront sensing is a new technique of high-resolution imaging through atmospheric turbulence developed at ONERA. The capabilities of the method have been demonstrated during first laboratory experiments. A deconvolution experimental set-up was installed on the 4.2 m William Herschel telescope on La Palma in November 1990. The results obtained during these observations are presented: the astronomical sources were observed at optical wavelengths; the resolution improvement reached a factor greater than 10.

The thermal emission phenomena in aerosol clouds is strongly related to the thermo-optical properties of each particle in the cloud. For this reason, we have studied the optical properties of a single particle, using the principle of radiative pressure levitation, with an Ar (or CO2) Laser. The particles have a diameter in the 5-80 micrometers range, and are heated at high temperatures when illuminated by a laser beam. These temperatures have been determined by optical pyrometry, on glass particles coated with aluminum or gold; radiance temperature up to 1000 K or more have been obtained. The general applications of this study can be found in the field of radiative heat transfer in liquid or solid rocket particles.