STRATIGRAPHIC studies in the snow of Antarctica have been directed at determining annual accumulation and defining factors which produce stratigraphie horizonation (e.g. Reference MellorMellor, 1964;Reference CraryCrary, 1971). Low, but highly variable accumulation rates, the redistribution of snow by winds, and shifting surface topography complicate the use of stratigraphie methods in Antarctica. Accumulation is generally attributed to snowfall, and Reference Taylor and CraryTaylor (1971) attributed the formation of stratigraphie reference horizons to metamorpbic processes rather than a reflection of variations in the originally deposited snow.Reference BensonBenson (1961) indicated the importance of deposition and diagenesis as a continuum in the formation of recognizable stratigraphie sequences. Recent observations on the Ross Ice Shelf suggest that rime and surface hoar make a significant contribution to annual accumulation and may produce horizons in the snow-pack which are difficult to distinguish from some depth-hoar layers.
Both ice and supercooled fogs are common in Antarctica. Table I summarizes some observations of the occurrence of fog on the Ross Ice Shelf for various years since 1929. Monthly values at Little America ranged from 3% for June to 20% for April and 24% for December (Reference Grimminger and HainesGrimminger and Haines, 1939) fogs are also a frequent occurrence in the Antarctic interior (Reference Astapenko.Astapenko, 1960}.
References: Λ—Reference Bilello and BatesBilello and Bates (1975); B-Reference Grimminger and HainesGrimminger and Haines (1939); C— Reference Hofmann, Hofmann, Dorrer, Nottarp. and MellorHofmann and others ( 1964); D—This note.
Rime and surface-hoar deposition accompanied all fog events at Camp C-7, and surface hoar grew on three occasions when there was no clearly visible fog. Fog most commonly occurred with light (0-5 m/s) south-easterly winds. Sustained snowfall accompanied the fogs when the winds were from the north. The fogs were clearly cellular in character, being marked by frequent temperature oscillations (as rapid as 0.5 deg/min), involving changes of up to 5 deg.
Nine measurements of the mass being added to the surface by rime and hoar were made on three separate days by measuring the volume of melt water from deposition on one meter square sheets of clear plastic which had been spread on the snow. This method yields minimum values because:
-
(a) Rime and hoar accumulated preferentially on the rougher than on the smoother natural surfaces and less collected on the plastic than on the surrounding snow;
-
(b) There was a small loss by evaporation during the melting of the samples; and
-
(c) Not all of the melt water flowed from the plastic into the graduated cylinder.
The first factor is the most significant and is estimated to produce undermeasurement of the mass being added to the snow surface by a factor of two or more.
Measurements made at C-7 ranged from 0.5 to 2.0 mg/cm2h, with a mean of 1.0 mg/2h. If fog and associated deposition of rime and hoar occur 10% of the time at any given location on the ice shelf (see Table I), and if the rate measurements taken at C-7 are typical, then an estimate of the minimum mass addition through this mechanism is 1 g/cm2 year. Limitations in the measurement technique suggest that the actual addition may be twice that. With an annual accumulation of 18.7 g/cm2 year at C-7 (Reference Crary, Crary, Robinson, Bennett and BoydCrary and others, 1962), rime and hoar deposition would comprise 5-10%, of the total. Using air humidity and air and surface temperature data to calculate the magnitude of moisture exchange between the air and the surface, Reference RusinRusin (1959) estimated that rime contributes 10-15% of the annual accumulation over large portions of Antarctica.
Rime and hoar deposited at C-7 grew in distinct clusters and would often form a 2-3 cm thick layer (Fig. 1). The layer was usually either buried by a subsequent snowfall or redistributed to the small-scale topographic lows by gentle to moderate winds. The redeposited material commonly maintained its low density and was, at times, in excess of 5 cm thick. However, when subjected to high winds, the rime and hoar was quickly broken and compacted.
Twenty distinct stratigraphie horizons were visible in a pit 2 m deep excavated at C-7. Some of the low-density layers, especially those associated with crusts, were clearly the result of depth-hoar formation. Others, however, are thought to record the deposition of rime and/or hoar on the surface. Up to 10% of the profile could be interpreted to be such layers. Layers which contained a surface hoar component might be interpreted to represent depth hoar. Although ". . .the interpretation of snow strata is based more on the recognition of similar layered sequences than on the positive identification of specific layers" (Reference Benson and CraryBenson, 1971, p. 341), the potential existence of horizons which derive some of their primary characteristics from their mode of deposition and which may resemble depth-hoar layers is of particular interest because some stratigraphers do attach temporal significance to depth-hoar occurrence (Reference Shimizu and MellorShimizu, 1964;Reference Koerner and Crary.Koerner, 1971;Reference RundleRundle, 1971).
Satellite photographs of the Ross Ice Shelf are normally received on a regular basis, and fogs can generally be distinguished provided there is no higher cloud cover over the same area. This is often the case; hence, the surface areal extent of these fogs and the approximate temperature conditions associated with their existence could enable further estimates to be made of their contribution to the annual accumulation on the Ross Ice Shelf.
Acknowledgements
The authors thank Mr D. A. Trachte for his assistance in the field. This work was conducted as pan of the Ross Ice Shelf Project, under NSF Grant No. OPP73-05843-AOI.