The Recession of an Icelandic Glacier

Abstract

THE accumulation OF quantitative data concerning rates of glacier recession in Iceland began in 1930, and figures relating to a few laciers have been made available.1 However, there is no such information for Hagafellsjokull Eystri and Hagafellsjokull Vestri, which are located at the south-eastern corner of Langjokull, nor, so far as is known, have Icelandic workers in this field made any measurements in that area. This paper constitutes an attempt to deduce, from an examination of maps and aerial photographs of different date, quantitative data relating to the reces? sion of Hagafellsjokull Eystri (E. Glacier) throughout the period 1934-56 and also during certain subdivisions thereof. A quantitative assessment is also made of the mean annual rate of lowering of the glacier surface during the interval 1950-6, and the resulting figure would seem to contribute to our knowledge of the life-history of dirt cones, at least in this particular area. The Hagavatn overflow-channels: Hagavatn (Fig. 1) is a typical ice-dammed marginal lake which came into existence when the building of Lambahraun, a post-glacial but pre-historic lava cone, blocked the drainage along an older north-east to south-west valley. The lake is dammed along its north-eastern side by the E. Glacier and it has at different times overflowed across the Jarlhettur range through four separate channels. A map prepared by the Icelandic cartographer Gunnlaugsson in the summers of 1834 and 1835 2 shows that Hagavatn, having abandoned its first outlet (Mosaskard),3 was at that time draining through channel 2 (Fagridalur). Since the heights of the four channel entrances decrease successively from south to north, flooding of the lowlands to the south will occur when the snout of the E. Glacier is receding, but not when it is advancing. Floods resulting from outpouring through channel 3 (Leynifoss) occurred in 1884,4 19024 and 1929.4' 5? 6? 7 Thus, prior to 1884, the snout of the E. Glacier must have been receding, but this phase must have been followed by a period of advance since subsequent recession resulted in another flood through channel 3 in 1902. Likewise, the glacier snout must have advanced again after 1902, otherwise recession could not have occasioned a third flood through channel 3 in 1929. By 1939 the glacier had further receded to such an extent that on August 13 of that year channel 4 was uncovered. Consequently, there was another disastrous flood over parts of the southern lowlands,8 and on this occasion the partial draining of the lake exposed a fine series of varves, particularly near the new south-eastern shore of Hagavatn. 9 The precise dating of the Hagavatn floods has, therefore, already made it possible to reach generalized conclusions concerning the fluctuations of the E. Glacier,4 but no quantitative data have hitherto been available. The recession of the E. glacier: a quantitative assessment: The Hagavatn area was mapped on a scale of 1 -.20,000 by the Cambridge University Expedition of 1934,10 and in August 1950 it was re-mapped on a scale of 1 '.25,000 by the Durham University Exploration Society.11 Vertical air photographs of the area taken on 23 September 1945, at a scale of approximately 1 -.45,000, were made available to the authors through the courtesy of Hr. Agust Bodvarsson of Landmaelingar Islands, Reykjavik. One of the tasks undertaken by the British Schools Exploring Society's 1956 Expedition to central Iceland was the re-mapping of the Hagavatn area on a scale of 1:25,ooo.12 It was hoped that by transferring to this map the positions of the E. Glacier snout in 1934, 1945 and 1950, it would be possible to present a quantitative assessment of the mean rate of recession of this glacier throughout the period 1934-56, as well as during the separate subdivisions thereof.