The surface structure and morphpology of the Soler Glacier, an outlet glacier located on the eastern side of the Patagonia Northern Icefield, were mapped and analysed, using near-vertical, aerial photographs, taken with a hand-held, 6x6 cm-format camera, in January, 1984. The major findings, which could hardly be obtained in the field, included identification of the five ice bodies and four sets of ogive patterns. The pattern and distribution of crevasses were also analysed. From the number and spacings of the ogives, the mean, annual, surface-flow rate was deduced to be about 100 m and 350 m, at points 1.4 km and 6.6 km from the snout, respectively, with an average of 160 m for the middle reach.

Since 1977, large areas on western Vatnajökull have been surveyed by ground-based, radio echo-sounding and the whole ice cap, HofsjökuIl, was surveyed in 1983. Detailed maps of the glacier-surface elevation and the sub-ice bedrock have been compiled. The instrumentation includes a 2–5 MHz, mono-pulse echo-sounder, for continuous profiling, a satellite geoceiver and Loran-C equipment, for navigation, and a precision pressure altimeter. The maps of western Vatnajökull cover about 1500 km2 and are compiled from 1500 km-long sounding lines, which yielded about 50 000 data points for ice thickness and 20 000 points for ice-surface elevation. The maps of HofsjökuIl cover 923 km2, the sounding lines were 1350 km long; 42 000 points were used for determining ice thickness and 30 000 for surface elevation. The maps obtained from these data are the first ones of the ice caps with surface elevation of known accuracy. The bedrock map of western Vatnajökull shows details of volcanic ridges and subglacial valleys, running north-east to south-west, as well as the central, volcanic complexes, Hamarinn, Bárdarbunga, and Grimsvtön and the related fissure swarms. The map of Hofsjökull reveals a large volcanic complex, with a 650 m deep caldera. The landforms in southern Hofsjökull are predominantly aligned from north to south, but those in the northern ice cap run north by 25° east.

Three ice-drainage basins on western Vatnajökull have been delineated with the aid of the first available, accurate ice-surface maps. These basins drain ice towards the major river systems: Tungnaá, Sylgja, and Kaldakvísl, There seems to be an important difference in the location of ice-drainage basins and melt water-drainage basins for the rivers Tungnaá and Sylgja. This is due to the influence of the bedrock topography on the flow of basal water.

Aerial photogrammetric block triangulation, a standard and well-developed technique for extending accurate control for mapping into the interior of a region from a few points of known position on its perimeter, can be readily adapted to determine surface velocities on bodies of ice which are too large, and often too crevassed, to be studied effectively by conventional ground surveying. Velocities are calculated from the changes in positions of the same natural surface features determined from photography of two (or more) epochs and the elapsed time. This method is capable of providing many uniformly-spaced measurements over the whole, moving, ice surface, thus allowing the production of maps of velocity and strain-rate, which are valuable in analyzing the ice-flow regime. Results from measurements completed some years ago on Byrd Glacier, one of the largest outlet glaciers from the East Antarctic plateau, are presented as an example of what the method can yield. By means of Doppler satellite surveying, relative positions of control points for each photography epoch can be determined with sub-meter accuracy, making the technique suitable also in regions where no fixed land features exist. A brief description of a project under way in such an area, on Ice Stream B in West Antarctica, is given.

The inventory of Glacier National Park, B.C., has provided an opportunity to test a number of new procedures which might be applied to future Canadian glacier inventories. The objectives of this particular study were to complete a tripartite inventory based on available map and photo sources and develop an automated system of data acquisition and processing. The conjunction of the two has permitted an analysis of the evolution of the glaciers in this area during the last 100 years. The techniques used and a summary of the results are reported.

The Illecillewaet Glacier (British Columbia) has been surveyed in a variety of ways since the end of the 19th century. A summary of the different information - archive photographs, field surveys, photo-interpretation, and remote sensing - reveals that between 1887 and 1962, the glacier had retreated more than one kilometre. However, since 1972, it has advanced about 100 m. The techniques used and the results are discussed.

For use in glaciological research, between 1982 and 1984, we succeeded in surveying and mapping the Mount Gongga Glacier, on a scale of 1:25 000, by means of a combination of terrestrial and aerial photogrammetry. This paper describes the method in detail.

In the survey area, we set up an independent, triangulation network, with microwave distance measurement, and two, independent, straight-line traverses, for basic control. Control points were observed by intersection. The terrestrial, photogrammetric baselines were projected and corrected into distances on the. plane of the map.

Terrestrial photography accounted for the majority of the photographs of the survey area. Surveying and mapping of planimetrie and topographic features were completed on a stereo-autograph, using plates mainly from terrestrial photogrammetry. Where these data were insufficient, they were supplemented by aerial photography, plotted on a photographic plotting instrument. Orientation points of the aerial photographs were established by terrestrial, photogrammetric analysis and located on the map by an optical, mechanical method.

The practical result showed that a combination of terrestrial and aerial photogrammetry, in mapping a high, mountain, glacier area, on a large scale, is more feasible and flexible than other methods and more economical as well.

Since 1981, the British Antarctic Survey has flown 27 000 line kilometres over Ronne Ice Shelf, during which radio-echo ice thicknesses were recorded. An earlier map of ice thickness was compiled from about 5000 km of flight lines flown by three different organisations over a period of 12 years A new, detailed map of the thickness of Ronne Ice Shelf has been produced, which agrees with the earlier version, but shows more structure in the north-western part. In particular, major ice streams maintain their idenity from sources such as Evans and Rutford ice streams all the way to the ice front. Regions where the ice shelf is locally grounded over Kershaw Ice Rumples and over more extensive ice rumples between Korff and Henry ice rises play a significant role in controlling the dynamics of the ice shelf. There is a possibility of extensive freezing-on of sea water under the thin, central section of Ronne Ice Shelf, although the radio-echo evidence for a thick layer of saline ice is ambiguous.

A statistical method to determine surface velocities from theodolite measurements is described. The method assumes that the measuring points move according to a model and the model parameters are estimated directly from the directional observations. Observations from two fixed points need not be simultaneous, which is an advantage in the case of fast moving glaciers. The method is practised on data from the Daugaard-Jensen glacier, in East Greenland, where the motion of the measuring points is calculated to the accuracy of the directional observations.

Landsat photographic products on the Space Oblique Mercator (SOM) projection are used to construct a map of Nordaustlandet (Svalbard), of known accuracy. The map includes ice divides. Accurately enlarged Landsat images were digitized. Combined digitizer and operator errors were 64 m, at the enlargement scale. Fifteen ground control points rixed the two scenes. RMS errors in control point identification were <123 m. Geographical coordinates were extracted by: 1) converting digitizer coordinates to SOM cartesian coordinates and 2) transforming these coordinates to latitude and longitude. This map production method is applicable to any imagery of known projection. The digitally stored map may be plotted on a variety of map projections and scales. Two problems in image interpretation were: I) shadows obscuring detail on NNE-facing coasts and 2) summer snow cover obscuring parts of the terrestrial ice cap margins, The map is similar to an east coast map produced from Landsat computer compatible tapes. Differences between the Landsat map and a 1:50 000-scale aerial photograph were <100 m.

Airborne geophysical investigations of the previously tittle-studied Nordaustlandet ice caps (11 150 km2) took place in 1983, using SPRI 60 MHz radio echo-sounding (RES) equipment of 160 dB system performance. RES and navigational data were recorded digitally. Navigation used a ranging system (accurate to ±30 m) from aircraft to ground-based transponders, located by satellite geoceivers, supplemented by the aircraft’s navigational instruments and timed crossings of known features. Ice surface and bedrock elevations were measured, using aircraft pressure altitude, terrain clearance, and ice thickness data. The mean error of 251 crossing points on Austfonna was 11 m. The reduced geophysical data are stored on a direct-access computer database. During 3400 km of flying, Austfonna (8105 km2) was covered by traverses a nominal 5 km apart, whereas a 15 km-spaced grid was flown over Vestfonna (2510 km2). Maps of ice surface morphology and subglacial, bedrock topography were produced for Austfonna and Vestfonna, along with an ice thickness map of Austfonna, Austfonna reaches a maximum surface elevation of 791 m and ice thickness of 583 m. 28% of the bedrock area beneath Austfonna lies below sea level. RES yielded bedrock echoes for 91% of track over Austfonna, but only 52% over Vestfonna. This was probably due to warmer conditions on Vestfonna, resulting in greater absorption and scattering of electro-magnetic energy. Ice surface elevations are a principal data source in the revision of official Norwegian maps of Nordaustlandet.

During the 1980 eruption of Mount St. Helens, the occurrence of floods and mudflows made apparent a need to assess mudflow hazards on other Cascade volcanoes. A basic requirement for such analysis is information about the volume and distribution of snow and ice on these volcanoes.

An analysis was made of the volume-estimation methods developed by previous authors and a volume- estimation method was developed for use in the Cascade Range. A radio echo-sounder, carried in a backpack, was used to make point measurements of ice thickness on major glaciers of four Cascade volcanoes (Mount Rainier, Washington; Mount Hood and the Three Sisters, Oregon; and Mount Shasta, California), These data were used to generate ice-thickness maps and bedrock topographic maps for developing and testing volume-estimation methods. Subsequently, the methods were applied to the unmeasured glaciers on those mountains and, as a test of the geographical extent of applicability, to glaciers beyond the Cascades having measured volumes.

Two empirical relationships were required in order to predict volumes for all the glaciers. Generally, for glaciers less than 2.6 km in length, volume was found to be estimated best by using glacier area, raised to a power. For longer glaciers, volume was found to be estimated best by using a power law relationship, including slope and shear stress. The necessary variables can be estimated from topographic maps and aerial photographs.

Grasshopper Glacier is a cirque glacier in the central Rocky Mountains of the United States. It is a remnant of the “Little Ice Age”, rather than the more widespread and older Pinedale Glaciation. The glacier has not been monitored on a regular basis and very few maps have been published of the area, but it has been studied, photographed, occasionally mapped, and described by scientific and non-scientific groups, at different times since 1898. These photographic, cartographic, and written records make it possible to trace the fluctuations of this glacier since 1898. Grasshopper Glacier has had periods of positive mass balance, but the overall trend has been negative, with accelerated melting in recent years. It is estimated that Grasshopper Glacier has lost about 50% of its area and as much as 90% of its volume, since 1898. Other Rocky Mountain glaciers are experiencing similar wastage and, if current conditions continue, these glaciers will disappear by the middle of the next century.

The Seasat radar altimeter observations of a 100 km2 area in South Greenland are compared to a detailed, ground-based survey, using “geoceivers” and pressure altimeters. The comparison shows the Seasat measurement of distance between satellite and earth to be accurate to the level of the geoceiver determined surface (±2 m). Due to the great distance between satellite and surface, finer details of surface topography are not revealed in the satellite measurements. As the satellite tends to lock onto hills in the vicinity of the sub-satellite track, the satellite tends to overestimate the true surface elevation. However, a similar altimeter would make a similar overestimate, allowing accurate differential measurement of volume changes between the two surveys.

Annual mass-balance measurements have been made at a number of glaciers in Norway since the beginning of the 1960s, A detailed and reliable map is necessary as a base for field work and more than twenty glacier maps have been constructed photogrammetrically at scales of 1:10 000 or 1:20 000 since 1952. For some of the glaciers more than one map has been constructed and changes in glacier volume can be calculated, provided the maps have sufficient accuracy.

For the glaciers, Nigardsbreen, Hellstugubreen, and Gråsubreen, two or more good maps are available and these form a good basis for comparisons and calculations of changes in volume between the years when the air photographs were taken.

Comparisons have been made for Gråsubreen between 1968 and 1984, for Hellstugubreen between 1968 and 1980, and for Nigardsbreen between 1964 and 1984.

The calculations are made by placing a 100 m grid on the glacier maps and comparing the altitude for corresponding points. The changes in height are regarded as representative for the 0.01 km2 glacier areas represented by each point.

Results from the investigation have been used to check the accuracy in cumulative mass balance for corresponding periods.

Repeated air photography, at intervals of 5—10 years, can be used in the future to find the cumulative mass balance for a great number of glaciers at lower cost than “normal” mass-balance work.

Techniques for mapping the distribution of buried glacier ice are discussed and the results, from a study carried out within the framework of flood protection work in the Italian Alps, are presented. Bottom temperatures of the winter snow cover (BTS) primarily indicate the heat flow conditions in the underlying ground and mainly depend on the presence or absence of an ice layer beneath the surface. Determination of BTS values is therefore an inexpensive method for quickly mapping the near-surface underground ice in areas where there is 1 m or more of winter snow cover. At greater depths, and/or when more detail is required, geoelectrical resistivity soundings and seismic refraction soundings are most commonly used to investigate underground ice. A combination of the two sounding techniques allows the vertical extent and the main characteristics (frozen ground, dead glacier ice) to be determined in at least a semi-quantitative way. Complications mainly arise from irregularity in the horizontal extension of the studied underground ice bodies, and they may have to be overcome by expensive core drillings and borehole measurements. Widespread occurrence of buried glacier ice was observed in morainic deposits, surrounding an ice-dammed lake near Macugnaga, Italy.