This paper reviews the electrical properties of snow and ice that are of importance in remote sensing using electrical devices. After a review of the observed laboratory behaviour of ice samples and the microscopic theory which has been advanced to explain this, the data on temperate and polar glacier ice are compared with the laboratory data. Temperate glacier ice is generally rather similar to laboratory ice, but certain relaxation processes found in the laboratory are absent from the glacier ice. Polar ice, on the other hand, is considerably different in its dielectric behaviour from "pure" laboratory ice, or temperate glacier ice; in many ways it more resembles doped laboratory ice, despite its variable, sometimes low, impurity content. It also resembles in behaviour ice produced by freezing supercooled water. The electrical behaviour of snow, and the attempts to account for this in terms of the behaviour of the ice and air components, and also of the water component in wet snow, are next discussed. Finally the implications of this work for radio-echo sounding of ice, radar reflectivity from wet and dry hydrometeors, devices for determining the water content of snow, and resistivity surveys of glaciers are discussed.

Two 0.5 m cores from "Byrd" station, Antarctica have been studied in the laboratory, one from a shallow depth (155 m) and the other from the zone where recrystallization has given a vertical c-axis fabric, and the air in situ is thought to be in clathrate form (1 424 m). The dielectric response has been studied in the frequency range 60 Hz to 10 kHz, and in the temperature range — 6° C to —6o° C. The behaviour observed is markedly different from that of "pure" polycrystalline ice such as may be made by slowly freezing distilled de-ionized water and is thus at variance with the conclusions of Rogers (unpublished) who deduced, from measurements of the admittance of a dipole probe lowered through the fluid-filled drill hole at "Byrd", that the ice surrounding the hole had a dielectric response similar to that of "pure" ice. The Antarctic ice is shown to have properties similar to those of the ice from "Camp Century" and "Site 2" in Greenland studied by Paren (1973). In an attempt to discover what factors determine the difference in electrical behaviour between polar ice and pure ice, some samples were melted and subsequently refrozen slowly. Their dielectric response was similar to that of pure polycrystalline ice. These results are discussed in connection with the impurity content and growth conditions of the ices.

After a brief review of factors relevant to the propagation of radio waves in ice, problems of profiling the upper and bedrock surfaces of ice sheets are considered. These include deconvolution of echo traces and accurate navigation. Results already available not only provide knowledge of sub-ice geomorphology, but also confirm our general ideas on interaction of temperature and flow in large ice sheets. Difficulties encountered in the study of temperate glaciers appear to be due to irregular dielectric properties of the ice mass, which in turn suggest an irregular distribution of water. In the future, application of radio-echo methods should lead to more detailed knowledge of accumulation and ablation over ice masses, improve our understanding of temperatures at great depths in polar ice sheets, and give data on ice movement from central areas of ice sheets.

We review the theory of interpreting the echoes received back at the source from a glacier bed illuminated with a wide-angle beam of quasimonochromatic radio waves. It is convenient wherever possible to separate bed relief into “geography” and “roughness”. “Geography” is the surface that results from horizontally smoothing the bed over distances comparable with the mean height h of the source above the bed. The geography can be deduced from the lime delays of specular reflections, provided these can be identified, “Roughness” is the statistics of the departures of the actual bed from the geography on scales down to about one-tenth of the predominant wavelength in the travelling pulse. The average height S and horizontal dimension L (where these quantities can be defined) of the irregularities can be deduced from measurements of the average value of the echo envelope p as a function of time delay. It is also possible to distinguish qualitatively different beds with the same S and L. If the source-receiver unit moves keeping h constant, then p fluctuates about its mean value. This “spatial fading” can be used to detect horizontal ice displacements of less than one hundredth of a wavelength. To detect vertical ice motion it is necessary to observe the phase of the echo.

Qualitative inspection of the results of the 1971/72 S.P.R.I.-N.S.F. Antarctic radio-echo sounding programme shows that it is possible to infer some characteristics of the lower face of the ice from the form of received echoes. We confirm the existence of lakes of liquid water beneath the east Antarctic ice by inspection of the top and bottom surface gradients of the ice, and suggest that basal melting occurs over a wide area in this region.

Quantitative studies in Devon Island indicate that small-scale irregularities in the bedrock are characterized by slopes of about 1: 40, with some higher gradients present, possibly indicating the presence of morainal boulders. We deduce that a geological boundary was crossed between 25 and 40 km west of the base camp, the rock to the west of the boundary having lower permittivity, and higher surface slopes than that to the east.

The question of the usefulness of the spatial fading of the echo in deducing surface characteristics is briefly discussed. We conclude that measurements of the fading are indeed useful, especially in the case of echoes with short “tails”.

An attempt to measure the surface velocity of a glacier by observing radio-echo fading patterns appears to have yielded instead a measure of the bottom sliding velocity.

A multitude of extensive layers have been observed by radio-echo soundings in Greenland. By comparison with the stable isotope profile from Camp Century it is found that layering in the lop of the ice has been formed in the period since the last glaciation. Radio-echo layers observed at greater depths in central Greenland may have been created in the period of the interstadials and further down possibly in the period prior to the last glaciation. Further investigations are needed to prove this.

Several seasons of aerial ice-thickness soundings over the region of the Prince Charles Mountains, the Lambert Glacier system, the Amery Ice Shelf, and their drainage basin in east Antarctica have now been completed. The measurements provide detailed maps of surface topography and ice thickness over an area of about 2 X 105 km2. The equipment used consisted of a 100 MHz echo sounder designed and constructed by Antarctic Division and carried in a Pilatus Porter aircraft. ERTS imagery provides a valuable background for portraying the echo-sounding results. These results show that an extensive, deep subglacial valley system forms the basis of the large drainage basin with concave ice surface topography which channels the ice flow into the Amery Ice Shelf. Deep glacial streams penetrate a long way into the ice-sheet basin. The rock relief is considerable, varying from 3 000 m above (present) sea-level to 2 000 m below sea-level. A very deep subglacial trench exists in the region of the confluence of the Fisher, Mellor, and Lambert Glaciers where the ice thickness reaches 2 500 m. The low surface slope and high ice velocity are suggestive of high melt production in this region. The strong echo, together with the high bedrock back-slope, suggests that the deep trench may contain a basal melt lake.

The techniques of geophysical exploration can be used in a number of ways to investigate the internal characteristics of ice bodies. The application of radio-echo sounding is discussed elsewhere by Robin (1975[a]) so is not considered here except for a few special cases. Anisotropy resulting from non-random ordering of crystal axes produces effects on seismic wave propagation speeds measurable by both refraction and reflection techniques. The effects on shear waves are particularly strong, so the recent development of an effective shear-wave generator should prove very useful. Marked effects on radio-wave polarizations have also been noted, but are not readily interpretable because the anisotropic dielectric characteristics of single-crystal ice in the radio-echo frequency range are not yet known. Elastic internal friction can be determined by careful measurements of seismic-wave amplitudes. Dielectric properties of the ice can be found from VLF propagation experiments and electrical logging in drill holes as well as from the radio-echo sounding techniques discussed by Robin (1975[a]). An activation energy can be estimated from d.c. resistivity measurements where the temperature profile is known. Recent developments have led to much improved methods of determining density-depth variations and long-term mean annual snow accumulation rates from seismic refraction shooting. Internal discontinuities which can be studied include a probable morainal layer detected by seismic reflections, a brine-soaked zone in shelf ice observed from radio-echo. sounding and electrical resistivity surveying, and inverted crevasses penetrating upward from the base of shelf ice, detected by radio echoes. Secular changes in gravity offer a sensitive means of determining long-term changes in surface elevation.

The errors involved in ice thickness determinations in Antarctica by seismic reflection shooting, gravity observations and radio-echo sounding are briefly discussed. Relative accuracies of 3%, 7-10% and 1.5% have been suggested. Double checks of ice depths from radar sounding in east Antarctica indicate an internal consistency of measurement for this technique of <1%. Comparison of carefully executed seismic shooting and routine radio-echo sounding results against absolute ice thickness values from two deep core drilling sites show no significant differences between these two remote methods (i.e. both are better than 1.5%).

Over 60 comparisons are examined between radar ice thicknesses and over-snow measurements obtained on eight independent traverses in east Antarctica. Three traverses exhibit consistently unacceptable results-U.S. Victoria Land Traverse II (southern leg), Commonwealth Transanlarctic Expedition and the U.S.S.R. Vostok to South Pole Traverse—which probably result from misinterpretation of “noisy” seismograms. The remaining comparisons indicate mean differences, including some navigational uncertainty, of ≈3%, <8% and 5% between radio-echo and (1) seismic, (2) gravity, and (3) gravity tied to seismic determinations, respectively.

Radio waves of fixed frequency, which had passed through ice and firn from a transmitting antenna lowered down a bore hole, were monitored on the surface. By mixing the received signal with the fixed frequency, the beats which were produced as the antenna was lowered gave a value for the wavelength in ice at the depth of the antenna. This was multiplied by the frequency to give the velocity of radio waves m ice. Measurements averaged over 10 m and longer intervals were related to density values from bore holes in the same location. Results above 50 m were difficult to interpret, but at greater depths the velocities fitted a linear relationship between refractive index and density. A small extrapolation gives a velocity of radio waves in pure ice of 167.7±0.3 /μs at — 20° C.

Many authors have used propagation of ultrasonic waves in ice for glaciological studies. This propagation is characterized by the velocity of sound and by the attenuation of stress waves. In crystalline materials, these two characteristics depend on structural slate. In particular plastic deformation gives velocity and attenuation variations.

We have measured the sound velocity and attenuation of ultrasonic waves in strained specimens of ice (single crystals and polycrystals). These measurements done between 100 and 273 K at a frequency of 5 MHz show that plastic deformation leads to an increase of attenuation arid an increase of velocity. Annealing treatments at 271 K cause recovery of propagation characteristics. The variation in attenuation can be interpreted by the theory of dislocations and this interpretation is supported by our data on the influence of frequency on this increase of attenuation induced by plastic deformation, but the theory of dislocations implies a decrease of modulus, i.e. of velocity, hence we must postulate that an added phenomenon screens the effect of dislocations. That phenomenon could be connected with ageing effects observed on different physical properties of ire and may be due to modification of protonic arrangement or creation of interstitials during plastic deformation. Thus our experiments show that it is necessary to be careful in using results determined from the propagation of ultrasonic waves in ice.

This paper presents an overview of the active microwave tools becoming available to the glaciologist with emphasis on recent radar developments as applied to floating ice. Sufficient theory is presented for the user to understand the techniques. Side-looking radar imagery is discussed using a number of examples resulting from the use of real and synthetic aperture, single and dual polarization. Recent studies of the microwave properties of ice and snow are reviewed, and are shown to be leading to significant advances in high-resolution radar techniques for accurate sounding of these materials. Remote sensing of fresh-water ice thickness is shown to be well established and operational, with similar techniques feasible in the near future for sea ice. It is pointed out that both imaging and probing radars applied to studies of sea ice and snow usually must be used in association with data from other sensors.

SLAR imagery of Nares Strait was obtained on three flights carried out in. January, March, and August of 1973 by Canadian Forces Maritime Proving and Evaluation Unit in an Argus aircraft equipped with a Motorola APS-94D SLAR; the March flight also covered two lines in the Arctic Ocean, from Alert 10 the North Pole and from the Pole down the long. 4ºE. meridian to the ice edge at about lat. 80º N. No observations on the ground were possible, but -some back-up was available on all flights from visual observations recorded in the air, and on the March flight from infrared line-scan and vertical photography.

The interpretation of ice features from the SLAR imagery is discussed, and the conclusion reached that in spite of certain ambiguities the technique has great potential which will increase with improving resolution, Extent of coverage per distance flown and independence of light and cloud conditions make it unique among airborne sensors.

To observe the distribution of pack ice off the coast of the Okhotsk Sea coast of Hokkaido, a radar network consisting of three radar stations was constructed during 1967-69. It covers an area about 70 km wide and 250 km long. The stations are remote-controlled by radio from the Sea Ice Research Laboratory and the information obtained is transmitted back to the laboratory and observed there. Radar has the great advantage of being able to make continuous observations of ice. Usually several special features can be seen on the radar screen, and they are used as markers for the observation of movement. It is ascertained that the average pattern of drift in this area is from north to south-east along the coast line and the ice field undergoes internal deformation during its drift. To get some information on the surface topography of ice from A-scope radar, the intensity of echo signals is classified into 16 steps by computer. To obtain the movement of an ice field from the numerical radar information, a modified two-dimensional cross-correlation method was tested.

During December 1973, the Naval Oceanographie Offirc (NAVOCKANO) and the Naval Research Laboratory (NRL) conducted a joint remote-sensing experiment over the sea-ice fields off Scoresby Sound on the east coast of Greenland using NAVOCEANO’s RP3-A Birdseye aircraft, laser profiler, and infrared scanner, and NRL’s 19.34 and 31.0 GHz nadir-looking radiometers. The objectives of this mission were: (1) to develop skills for interpreting sea-ice passive microwave data. (2) to expand, if possible, the two-category capability (multi-year ice and first-year ice) of passive microwave sensors over sea ice, (3) to compare two frequencies (19 and 31 GHz) to determine which may be more useful in a scanning radiometer now under development at NRL, and (4) to determine the value of multi-frequency as compared to single-frequency study of sea ice.

Since, because of darkness and remoteness, no photography or in situ ground truth were possible for this mission, it was necessary to rely on the interpretations of the laser and infrared (IR) data to evaluate the performance of the microwave radiometers. Fortunately, excellent laser and IR data were collected, and a confident description of the ice overflown was possible.

Five ice conditions: (1) open water/new ice, (2) smooth first-year ice, (3) ridged first-year ice, (4) multi-year ice, and (5) a higher brightness temperature form of multi-year ice interpreted as second-year ice were identifiable, regardless of weather conditions, by comparing the average of the two microwave brightness temperatures at the two frequencies with their difference.

Ice and water surface temperatures were measured with an airborne radiation thermometer PRT-5 over the North Water polynya during three missions between late winter and early summer 1974. Error corrections, problems of data analyses and mapping are discussed. Attempts are made to relate the main types of sea ice to temperature ranges, which then are used in conjunction with satellite pictures to produce surface temperature maps.

This paper discusses the measurement of important snow properties using electromagnetic radiation. Snow areal extent can be measured using manual, optical, electronic, or digital techniques from data supplied by visible and near-visible light sensors carried on Earth resources and meteorological satellites, but these techniques cannot routinely detect snow under clouds or a forest canopy. Gamma-ray techniques used at stations or from low-flying aircraft permit measurement of water equivalent of snow (depth times density). Active or passive microwave systems may permit this to be done over larger areas, but the physics of this possible technique is not yet sufficiently understood. Wetness or temperature, of a snow surface is measurable with thermal infrared radiometers; wetness throughout a snow pack may be measurable with microwave radiometers. Much research needs to be done on the electrical (including scattering) properties of snow before efficient, all-weather, remote-sensing systems can be designed.