This paper reviews some general characteristics of atmospheric observations and numerical weather prediction methods which provide most of the fundamental data used by meteorologists providing operational environmental support. In general, these data have temporal, spatial, and informational shortfalls for providing real-time inputs for atmospheric transmission support. Meteorological models have improved significantly since first conceived back at the turn of the century and this trend is expected to continue. Improvements are a direct result of gains in the quantity and quality of observational data as well as rapid increases in computer technology which allow for more complete and higher resolution models. However, there still remain a number of deficiences with current observations and models that need to be considered by those concerned with electro-optical or other electro-magnetic transmission through the atmospheric environment.

Turbulence, atmospheric background, and aerosol forward scattering modulation transfer functions 0M Fs) are analyzed. For diagonal or vertical imaging turbulence is seen to limit image quality only at very high spatial frequencies, where degradation is likely to take place anyway as a result of vibration and diffraction. Background and aerosol MTFs limit law spatial frequency contrast as well. For long-range horizontal lines-of-sight near ground level, both turbulence and forward scattering by aerosols severely limit image quality. Numerical examples are presented. These limitations can be overcome somewhat by proper selection of the imaging wavelength and of operation timing. This analysis can aid in sensor selection for system design from the standpoints of both wavelength selection and sensor resolution. Because this analysis includes the effects of weather changes on image propagation through the atmosphere, and a model for predicting image quality as a function of weather forecast, it also can aid in selecting operation timing on the basis of weather forecasts, with a view toward optimizing expected resolution.

The goal of the paper is to show that the determination by optical remote sensing methods of reliable vertical concentration profiles or column densities necessitates the best possible knowledge of the spectroscopic line parmeters involved (positions, intensities, linewidths...). In the first part some recalls concerning molecular spectroscopy are given whereas the second part gives specific examples (isotopic ozone, nitrogen dioxide, deuterated water vapor) illustrating clearly the need of precise spectroscopic parameters in order to make reliable optical remote sensing measurements in the atmosphere.

Atmospheric emission and absorption infrared spectra obtained with several of the University of Denver's balloon-borne and ground-based spectrometers are presqnted. The spectra cover intervals of the 3-25 micron region with resolution ranging from 0.5 to 0.0005 cm^-1. Analysis of these spectra provides a powerful tool for the identification and quantification of atmospheric trace gases, and for updating spectroscopic parameters needed for modeling atmospheric radiation.

The objective of the ATMOS experiment is to measure the concentrations and distributions of gases in the middle and upper atmosphere (10 km to 120 km) by infrared absorption spectroscopy. In 1985, during NASA's Spacelab 3 mission, high resolution (0.01 cm^-1) infrared occultation spectra of the atmosphere were recorded using a modified Michelson interferometer orbiting on board the space shuttle. With the sun as the radiation source, a single spectrum was obtained every 2.2 seconds with signal to noise ratios of 100:1 or better. Many disciplines and technologies were combined to produce this state-of-the-art NASA experiment. An overview of the data analysis techniques, implications for molecular spectroscopy and results obtained from the SL3 mission is presented.

This paper reviews the development of the atmospheric molecular absorption parameters. These data are fundamental input for line-by-line computer codes that simulate atmospheric transmission or radiance. A review is given of the basic parameters required and the history of the molecular databases. Some illustrative examples are given of the source of the data. The directions that the database will take in the future is discussed.

A description is given of the basic elements of a line-by-line computer model with particular reference to a new transmittance/radiance code GENLN2 currently being developed. Calculations may be performed for a multi-layered atmosphere of mixed gases and for different viewing geometries. The main problems associated with line-by-line calculations are concerned with line shape and continuum absorption, the treatment of lines outside the given spectral range, and the way in which spectral sampling is performed. These are considered together with methods of improving speed of computation in general. Results and intercomparisons from different applications of the technique are presented.

An array of models and techniques exist for the calculation of the atmospheric backscattered radiance and ground reflected radiance received at satellite altitudes. In remote sensing applications, where one deals in mega-byte units of data, it is essential that these models be computationally fast while retaining a reasonable degree of accuracy. We have evaluated a number of such models (Lowtran 6, Turner, Discrete ordinates method, 5S) relative to an accurate multiple scattering, multi-layer (Dave) model in order to assess the performance of these models in an inversion scheme for aerosol optical depth. The Turner and single scatter Lowtran 6 models generally produced large errors in apparent reflectance. Overall, the Turner model was not significantly better than the Lowtran 6 model except at near nadir geometries and non zero albedos. The 5S model which is orders of magnitude more rapid than the DOM model was significantly more accurate than the L6 and Turner models. The accuracy of the inversion procedures for the extraction of aerosol optical depth from satellite apparent reflectance was then analyzed for the two most precise models (DOM and 5S). For typical measurement conditions, the 5S inversion errors were found to be of the order of .1 in a turbid atmosphere case (aerosol optical depth approximately .5). The DOM produced the most impressive results in terms of comparisons with the Dave computations. It's time of execution, however, is a serious constraint with respect to satellite remote sensing applications.

Modeling of atmospheric propagation seeks to provide functional relationships between path thermodynamic properties (such as temperature, pressure, density, and the concentrations of relevant absorbing and scattering gases and aerosols) and corresponding spectrally dependent, path optical properties (such as extinction, optical depth, and radiance). An equally challenging and practical problem can be posed. Given data on the spectral transmittance or emittance (i.e. radiance) of a specified path, what can be determined about the variability of its thermodynamic and compositional properties? These interdependent aspects of path characterization modeling are expressions of the classical forward and inverse problems of radiative transfer, respectively.

Optical turbulence is observed to cause beam wander, scintillation, small coherence lengths, limited bandwidths and isoplanatism constraints on the propagation of visible wavelength laser beams. Developers of laser systems and electro-optical systems need to know and employ the appropriate representations of turbulence in order to estimate their real system performance. Unfortunately, there exists only a limited number of measurements of optical turbulence. Hence it is crucial that the extent and limitation of current models of optical turbulence be understood. A review of past and present models that describe turbulence throughout the atmosphere is presented. The utility and limitation of these models is determined by comparison to various measurements. Models represent a compromise in representing averaged data and may never represent any specific single optical turbulence altitude profile.

Infrared radiation leaving the earth's atmosphere (Infrared airglow) is briefly described. The concept of non-local thermodynamic equilibrium is introduced and the sources of infrared airglow are briefly discussed. Airglow originating from various vibration-rotation bands of CO₂ is discussed to serve as an example for the concepts introduced.

The phenomenon of line mixing in infrared vibration-rotation spectra and its impact on atmospheric transmission and radiative transfer is discussed. Line mixing, which is often also termed line coupling, occurs when rotationally inelastic collisions cause transitions among overlapping spectral lines. These collisional transitions can produce, in extreme cases, an overall narrowing of the blended profile of the coupled lines and can dramat-ically lower the absorption in the wing of a vibration-rotation band. Laboratory spectra exhibiting line mixing in both Q-branch and R-branch spectra of CO₂ are examined. Methods for the calculation of line mixing in atmospheric spectra are discussed, including methods for dealing with temperature dependencies. Observations of line mixing in both limb and nadir viewing atmospheric spectra are presented.

The thermal infrared spectrum of water vapor beyond 8 pm consists of a fairly well-understood spectrum of rotational lines and an inadequately described continuum spectrum underlying the observed spectral lines. The contribution of the continuum absorption can be important in the calculations of the atmospheric radiative heating rates and of the net surface flux. This paper presents some new results on the continuum absorption in the thermal infrared between 8 and 18 μm along with a brief review of currently adapted empirical and semi-empirical models explaining its nature.