The winter environment can produce obscuration effects which can severely degrade the effective operation of visible, infrared, and millimeter wave systems. The phenomena responsible for these effects can be classified into two main categories pertaining to the atmosphere and the terrain. This paper summarizes data needs for characterizing obscuration effects in winter and includes examples of some cold regions equipment for obtaining the data. The impact of various winter optical phenomena on multispectral systems performance is discussed.

For several years radiation fog field programs have been conducted at Albany, NY, with an emphasis on understanding the basic mechanisms leading to dense fog formation. This past year a cooperative effort ("Fog Project-1982") involved nine university, federal and private research laboratories, including NCAR staff and their remote system of 25 portable automated mesonet (PAM) weather stations. A number of comprehensive data sets (boundary layer meteorology and cloud physics variables) during the 14-16 hour nocturnal evolution of fog have been obtained. In particular, the extinction of light in the visible and infrared (10.6 pm wavelength), associated visibility, drop size distributions, liquid water content, and vertical tethered-balloon soundings provided new insights into the structure of fog. A CO2 laser transmissometer was developed that yielded direct information on fog density. During October of 1981 and 1982, a number of radiation fogs occurred that were super-cooled in their lowest 20-50 m. This posed certain troublesome to critical measurement problems with several instruments. Cold environment techniques were devised to overcome some of these instrumentation difficulties.

The transmission of infrared and microwave radiation through falling snow is a subject of interest to the United States Army. The Cold Regions Research and Engineering Laboratories are currently studying the transmission characteristics of various wavelength systems through falling snow. A portable ice crystal replicator which operates unattended has been designed and built for use in this program. Of particular concern in its design was the problem of 'blushing', which results fram the freezing of water on the Formvar plastic material during the replication process. This problem has been at least partially solved in the design of this new instrument. The instrument also incorporates various electronic timing and event recording devices so that the instrument can be incorporated into a larger complex of instrumentation carrying out these transmission studies. A description will be given of the instrument and details presented of ice crystal characteristics that have been observed with the instrument under field conditions.

A system has been developed by the U.S. Army Cold Regions Research and Engineering Laboratory to measure the mass concentration of falling snow crystals. It is known as ASCME (Airborne Snow Concentration Measuring Equipment) and is described in this paper. ASCME's general performance has been evaluated based on concurrent measurements of precipitation rate. A strong correlation between airborne-snow mass concentration and precipitation rate yields an estimate of particle fall velocity (= 0.92 m/s) close to that observed by other researchers. Factors affecting system accuracy have been investigated and are discussed. Examples are given of the utilization of ASCME data in analyses of electromagnetic energy propagation in falling snow.

Three, newly-developed snow characterization instruments are described together with examples of their data products. A snow rate meter was developed that utilizes a sensitive electronic balance to determine snow rate to an accuracy of about ± .05 mm/hr (water equivalent rate) at time resolutions varying from some 15 to 120 s, depending on the wind turbulence situation. A fall velocity indicator was constructed utilizing vidicon TV recording and strobe illumination that permits the fall velocities and fall characteristics of snowflakes to be ascertained. A snow structure recorder was also designed that uses a vidicon camera to look at the detailed crystalline structure of snowflakes (over a 1 cm x 1 cm field of view) that fall on a sampling area which is an intermittently moving belt.

Snow as an obscurant presents some interesting challenges to those attempting to characterize it. The wide range of particle sizes which can he present at any instant, and the intricate and varied particle geometry, which makes particle orientation an important consideration in snow characterization and extinction measurements, both call for the use of special measurement techniques. The application of particle size spectrometer probes to the measurement of distributions and area concentrations for snow crystals and flakes in the 12.5- to 6200 -μM size range is described.

The design and operating characteristics are presented for a visible and near-infrared scanning photometer which can measure incident and reflected spectral irradiance from 400 nm to 2,450 nm. The instrument is designed to record over 98% of the solar radiation incident upon and reflected from the Earth's surface using a turret-type cosine collector and a circular variable interference filter. The apparatus has been tested successfully on arctic sea ice at temperatures down to -20°C. It is self-contained and easily portable, and its mass is less than 16 kg. Some preliminary results are presented.

Providing ground truth for the backscatter and absorption effects of a snow cover on electromagnetic waves has long been a problem. One characteristic of the snow cover which has been particularly difficult to measure is its free, or liquid, water content - the fraction of the snow's volume which exists in the liquid state. Five methods which have been used for measuring this parameter are described and their merits and deficiencies discussed. Two of the methods are calorimetric, measuring the free water content as a function of the heat added to or removed from a snow sample while completely melting or freezing it. The third uses the freezing point depression observed on adding a salt solution to a snow sample to calculate the snow's free water content. In the fourth procedure, a snow sample is completely dissolved in ethyl or methyl alcohol. The corresponding decrease in temperature is inversely related to the free water content of the snow. The final technique is electronic: above a certain frequency, the electrical capacitance of snow is related to its density and free water content. With accurate calibration, devices which measure snow capacitance are likely to be the simplest and fastest means of providing free water measurements.

Ice particles in the atmosphere range in size from ice fog particles a few pm diameter to hailstones some 10 cm across. Many smaller particles tend to have singular crystalline facets and grow in the form of either solid or hollow hexagonal columns or needles, hexagonal plates, sector plates or dendrites. Size and shape determine fall speed and orientation. Beyond a size of a few hundred microns, aggregation to snowflakes takes place and particles may rime, to form soft hail particles, and ultimately higher density hailstones; aggregation to snowflakes takes place. Particles often depart drastically from the idealized shapes and dimensions sometimes used in modelling radiation interactions. Particles may be characterized by scales downward from their maximum dimension, to surface steps and bubbles of dimension < μm. Size and fall velocity distributions are intimately related to generation and velocity sorting mechanisms, and may be quite narrow or very broad depending on circumstances.

Snow appears white under most conditions of illumination. Yet holes in snow are blue. Both are consequences of the absorption properties of pure ice and multiple scattering by ice grains. The albedo of pure snow is high and does not vary greatly over the visible spectrum because incident photons that are scattered many times before re-emerging from the snow are unlikely to be absorbed before doing so. It is only by observing transmitted light that colors are seen. Beautiful blue light is often seen in quite shallow holes in snow. Although it takes many meters of ice to selectively absorb visible light, a few tens of centimeters of snow are sufficient because multiple scattering increases the effective path length a photon travels before reaching a given depth. In contrast with snow, frozen waterfalls (bubbly ice) are sometimes observed to be slightly green or bluish-green. Moreover, they are not as highly reflecting as surrounding snow. A first approximation to radiative transfer in bubbly ice is to consider it to be snow with a very large grain size. For snow with grain sizes a few millimeters or less the albedo is high and does not vary appreciably over the visible spectrum. But if the grain size is increased to 10 millimeters or more the albedo drops markedly and differences over the visible spectrum become more pronounced.

This paper describes characteristics of visual range attenuation in blowing snow in rela-tion to motorist vision. Visual range V is related to wind speed U according to V = A U-5 with the A coefficient changing in response to snow availability. A lower limit for visual range is described with A = 1.1.108 m6.s-5 An operational monitoring system used on Interstate Highway 80 in Wyoming demonstrates how real-time computer analysis of photometric data can be used to determine the A value, interpret visual range in terms of vehicle operation, and provide automated traffic operations decisions.

Visible and infrared extinction in falling snow as observed by conventional transmissometers has shown a spectral dependence such that extinction increases with increasing wavelength in the absence of coexisting fog. Explanations of this wavelength dependence are given and techniques for modeling visible and infrared transmittance through falling snow are discussed. The spectral dependence may be used in the remote sensing of mass concentration and precipitation rate of falling snow.

A method is presented for extracting corrected extinction coefficients from transmittance measurements in a scattering atmosphere, by computing the unscattered, first-order scattered, and second-order scattered contributions to received power. These computations are based on the optical phase function of the scattering particles and on the optical characteristics of the measuring system. The method is applied to transmittance measurements in falling snow, which were made simultaneously at 0.6328, 1.06, 3 to 5, and 8 to 12 micrometers by using two laser transmissometers and an infrared transmissometer on a one kilometer horizontal outdoor path. The optical phase function of the snow crystals was measured at the same time as the snow transmittance measurements, by using a polar nephelometer with a 0.6328 micrometer light source. Models based on geometrical optics was then fitted to the measured data to generate a continuous function from 0 to 180 degrees, and this model was also used to extend the phase function to 1.06, 3 to 5, and 8 to 12 micro-meters. Extinction coefficients extracted by this method from the 1.06, 3 to 5, and 8 to 12 micrometer transmittance data are compared to those at 0.6328 micrometers.

Snow characterization measurements obtained during SNOW-ONE-A and SNOW-ONE-B are presented. These consist of data that were acquired by the snowrate-meter, fall-velocity-indicator and snow-structure-recorder instruments. These instruments were previously described in the PSIE Sub-session on "Electro-Optical/Infrared Systems and Effects"(Plank, Matthews and Berthel). E/0 correlations between snow rate and extinction/transmittance are illustrated, as are the correlations between snow rate and liquid water content, which are related by the fall velocity of the snow crystals. Snow rate, fall velocity, and snow structure data are presented.

This paper reviews the results based on the Dolin-Fante method based on the small-angle approximation solutions of the radiative transfer equations for laser beam propagation in dense nonspherical particles; for example, snow. Numerical results will be given for multiple scattering effects as functions of the detector's field of view and aperture size for typical published snow scattering phase functions.

During the past 18 months the Wave Propagation Laboratory has operated a ground-based pulsed coherent lidar for remote sensing of atmospheric winds and backscatter. Under normal summer operating conditions, the lidar measures winds to vertical heights of 10 km or to horizontal ranges beyond 15 km. This measurement capability is degraded, however, by the presence of hydrometeors such as snow, rain, and water or ice fogs. Although presenting a much stronger target to the lidar, these phenomena absorb and scatter incident radiation to such an extent that the extinction coefficient is greatly increased. Data are presented that compare system performance during snow and fog with performance under normal operating conditions. Results of lidar parameter measurements from calibrated targets as well as atmospheric scattering are discussed.

While the propagation of visible and infrared radiation in adverse weather has been studied by many investigators, the effect of rain on contrast transmission or visual range have received little attention. A theoretical and experimental program examining contrast transmission in adverse weather will be described in this paper. Measurements of contrast transmission were made using bar targets consisting of parallel flourescent light bulbs. The bulb spacing is four times the bulb diameter. Data is obtained by photographing the target from a distance of two hundred meters using a camera fitted with an eight-hundred millimeter lens under both clear and obscured conditions. The negatives are then analyzed using a microdensitometer and contrast is defined by C = (I MAX - i MAX + i MIN)-1 where I-MAX and I-MIN are the intensity maximum and minimum, respectively. In addition to contrast measurements, simultaneous measurements of atmospheric transmission, snow particle phase function, and all relevant weather parameters are made. A computer model, based on scattering theory, has been developed for predicting contrast transmission. The model computes zero, first and second order scattering contributions in a three-dimensional atmosphere using phase function computations based on Mie theory.

A Fabry-Perot interferometer is used to observe the extinction cross-section of water droplets in the mm spectal range. The size ranges over that of interest in atmospheric propagation.

Atmospheric effects on the propagation of electromagnetic waves are usually described in terms of intensity fluctuations, angle-of-arrival fluctuations, and the mutual coherence function (HCF). Experimental determination of these statistical quantities is usually accomplished by forming a time average of instantaneous field measurements and assuming ergodicity to obtain an ensemble average. This paper reviews two experimental approaches used to obtain such data, viz., the long-baseline interferometric method and the quasi-optical method; emphasis will be placed on the latter method. In addition, sampling of the atmospheric temperature and humidity fluctuations, which give rise to the electromagnetic fluctuations, is analyzed. Finally, a 1.6-km propagation range employing the quasi-optical method and the meteorological sampling considerations mentioned above will be described. The range is completely instrumented to provide the relevant meteorological and electromagnetic para-meters needed to characterize propagation at frequencies near 300 (Hz in a turbulent atmos-phere.

During the SNOW-ONE and SNOW-ONE-A exercises at Camp Ethan Allen, Vermont, measurements of attenuation through falling snow were made at 35, 95, 140, and 217 GHz over a two-way path length of one kilometre. Attenuations measured appeared to be a function of crystal size, free water content and snow mass concentration. Instrumentation design and the results of measurements are presented with some general observations of turbulent phenomena.

In January and February of 1981 and 1982, the Ballistic Research Laboratory (BRL) participated in SNOW-ONE at Camp Ethan Allen, VT, sponsored by the US Army Cold Regions Research and Engineering Laboratory (CRREL). The BRL conducted high-angle radar measurements there, measuring backscatter at 35 GHz. from various areas of undisturbed snow, acquiring time series data under varied weather conditions. A dual-polarized, 35 GHz radar was used to make the snow backscatter measurements. Polarity of the transmitted signal, vertical or horizontal, was controlled by means of an R.F. switch. The received signal was passed through an orthomode transducer to two receivers, allowing both parallel and cross polarized components to be recorded simultaneously. The sensor was supported 15 meters above the ground on an elevation-over-azimuth antenna mount. Elevation angle was adjusted so that radar beam angle was 30-degrees from vertical. Azimuth angle was varied so as to scan areas of undisturbed snow around the base of the support. The sensor was enclosed in an insulated box and heated to a fixed temperature. Signal returned to the sensor by the snow was, in large part, dependent upon the condition of the surface snow as it was affected by air temperature. When the temperature was well below freezing, sigma-zero was around -7dB with parallel polarization (-13dB, cross polarization). When the temperature increased to the freezing point, the snow became wet, packed easily, and had a lower sigma-zero (-17dB, parallel, and -22dB, cross polarization). The condition of the snow surface had a secondary effect upon the value of sigma-zero. A smoothed surface reflected less energy back to the sensor than a roughened surface. The combined effect of these two variables produced the changes that were measured in sigma-zero.

Snow precipitation takes a variety of forms depending on the conditions in the atmosphere at the time of the snowfall. Regardless of what particular conditions prevail at that time, once the snow particles reach the ground they immediately begin changing. This is not surprising since the snow cover is at or close to its melting temperature, has a very large specific surface area, and has ever changing boundary conditions.

Cold regions present land, coastal and oceanic geographies to observers. While this paper focuses on polar regions, it must still consider the coastal and oceanic constituents that exist there. However, polar environments augment these similar geographies with some remarkable environmental conditions that provide unusual challenges to observers. This fact produces a need for special tools whether observations are made by men on the ice or they are made remotely from aircraft or satellite platforms. This paper considers the special needs and requirements of cold regions users and discusses how space based remote sensing has the potential to uniquely serve these needs. Some of the historic remote sensing programs and sensors that have contributed to polar observations are reviewed and the promise of future missions is evaluated with an eye toward accomplishments that hold the promise of significant commercial and related societal benefits in the coming decades.

Studies of the observing needs of a diverse set of space science and application disciplines from the late 1980's through the 1990's indicate that individual science questions require multiple sensors and that there is considerable overlap in the observations needed to answer different questions. Subsequent studies of the data acquisition requirements for these sensors also indicated a synergism to be gained from grouping of instruments with similar orbital and orientation requirements. Other studies have suggested that a cost-effective approach to implementing these requirements would be to group instrument sets on several large platforms with different orbital parameters. These space platforms would have large mounting areas capable of accommodating a number of large instruments and these instruments would share a common power supply, attitude controls, thermal control, communication link, etc. In addition, these space platforms could be serviced by the Space Shuttle, thereby providing a capability to replace or upgrade individual instruments or instrument sets, change out instruments at the focal plane of a large telescope, replenish consumables, and provide general maintenance and repair services.

The TM aboard Landsat-4 launched on July 16, 1982, represents a major advance in Earth resources sensors. Its seven spectral bands record surface radiation in blue, green, red, near infrared, middle infrared and thermal wavelengths. The spatial resolution of approximately 30 meters represents a sevenfold increase over the previous Landsat sensor, the multispectral scanner subsystem (MSS). In addition, TM has greater radiometric sensitivity, distinguishing 256 quantization levels, compared with 64 for the MSS. These potential improvements have significant implications for satellite remote sensing in cold environments. The addition of the middle infrared bands will permit clouds to be distinguished from snow. It may also be possible to relate spectral response in this range to snow condition and hence water content. The thermal band responds to differences in surface temperature, which may be related to variations in soil moisture and drainage. These are important considerations for cold region construction. Water features, which can be related to circulation and quality, can be detected in the thermal and blue bands as well as the previously available green and red. The finer spatial resolution of TM will permit more accurate location, differentiation and monitoring of the extent and movement of ice and snow. Both spectral and spatial characteristics of TM contribute to more accurate land cover classification and geographical location of individual land cover features. These improved capabilities will make the TM considerably more useful for both civilian and military needs in cold environments.

The earth as viewed from space in visible light at night reveals some features not easily discernible during the day such as aurora, forest fires, city lights and gas flares. In addition, those features having a high albedo such as snow and ice can be identified on many moonlit nights nearly as well as they can in sunlight. The Air Force DMSP satellites have been operating in the visible wavelengths at night since the mid 1960's. Most all other satellites having optical sensors are incapable of imaging at night. Imaging systems having improved light sensitivity in the visible portion of the spectrum should be considered when planning future earth resources satellite missions in order to utilize nighttime as well as daytime visual observations.

Photoanalysis of coastal areas can give ambiguous results when parameters such as mobility and trafficability need be determined. Multiband imaging radar can augment or replace photography for such applications. Preliminary evidence shows that L-band imaging radar will (1) penetrate most wetlands vegetation species to delineate obscured shorelines, and (2) discriminate varying levels of substrate saturation and thus, potentially, quantify varying degrees of trafficability. We know that X-band radar can map many species of wetlands vegetation with good accuracy. The combination of L- and X-bands should provide a significant breakthrough for wetlands reconnaissance and mapping.

Since 1977 the NASA Airborne Oceanographic Lidar (AOL) has been utilized to evaluate the potential of airborne lidar systems for a variety of marine and terrestrial applications. The AOL is designed as a flying laser laboratory with flexibility that allows rapid modification of transmitter and receiver optical configurations as well as operation with various lasers. This flexibility in design has permitted the use of the AOL for numerous types of investigations in differing and often unrelated disciplines. The AOL can can be operated in two basic modes; backscattered signals can be temporally resolved and recorded in the bathymetric mode, while in the fluorescensing mode returning on-wavelength, water Raman, and laser induced flourescence response signals are spectrally resolved. Results of investigations conducted during the past several years over marine and terrestrial targets are discussed along with planned improvements to the lidar system. Results are presented for terrain, shoreline, and ice topography, and hydrography performed in the bathymetric mode as well as for chlorophyll a and phytoplankton photopigment investigations performed in the fluorosensing mode.

Airborne measurements of the natural terrestrial gamma radiation flux near the soil surface are used to infer mean areal snow water equivalent and soil moisture values. Terrestrial gamma radiation data are sensed from a lowflying aircraft over a network of 400 flight lines (each approximately 6 km2) in the upper Midwest. The airborne data are used to make real-time airborne snow water equivalent and soil moisture measurements in the region to support the National Weather Service hydrologic forecasting program. Ground-based snow cover data collected along 20 calibration flight lines indicate that airborne snow water equivalent values can be calculated with an RMS error of 0.88 cm. Ground-based soil moisture data collected along 155 calibration flight lines indicate that airborne soil moisture values for the upper 20 cm can be calculated with an RMS error of 3.9 percent soil moisture. Approximately 80 percent of the error is due to airborne instrumentation sensitivity while only 20 percent of the error results from flight line calibration using groundbased soil moisture data.

January's launch of the Infrared Astronomical Satellite (IRS) put an imaging infrared telescope in space. Its subsequent performance gives after-the-fact justification for the choice of beryllium as the material for optics and structure. In this paper, we examine the IRAS pre-flight test data to uncover the causes for the projected diffraction limit of ten micrometers and we will show that there is reason to project a considerably shorter practical wavelength limit for future applications such as Shuttle Infrared Telescope (SIRTF).

The mirrors of the 40 cm GIRL telescope cooled by superfluid helium will consist of glass ceramic Zerodur. The poor thermal conductivity of this material called for unusual cooling procedures, the details of which have been investigated in laboratory tests. The assembly of the mirrors at the aluminium base occurs via an intermediate invar structure which is suited to maintain the optical quality after cooling. Special expansion compensators guarantee equal focussing conditions at laboratory and cryogenic temperatures. Vibration tests of the engineering model have proved the dimensional stability of critical components. The infrared reflectivity of different mirror desposits were measured under cold conditions.