Mapping ice front changes of Müller Ice Shelf, Antarctic Peninsula Antarctic Science 7 (2): 197- 1,05 ( 1 995) Short note Mapping ice front changes of Miiller Ice Shelf, Antarctic Peninsula CAROLINE G. WARD British Antarctic Survey, High Cross, Madingley Road, Cambridge, CB3 OET Present address: Coventry University, Priory Street, Coventry, CVl SFB Introduction Miiller Ice Shelf (67"15'S, 66'52'W) is situated at the southern end of Lallemand Fjord. It is a small ice shelf (- 80 km2) fed by Briickner and Antevs glaciers, which both flow northward off the central peaks ofArrowsmith Peninsula; the ice shelfcontains an ice rise (Humphreys Ice Rise). Data sources have indicated that not only has the ice front retreated since 1947 but also that there have been two advances. This paper describes how these changes were recorded using simple photogrammetric techniques. Methodology Data sources Seven sources of data were used (Table I), spanning the period 1947-1993. These included both oblique and vertical aerial photographs, and visible satellite imagery. The quality of the images used varied greatly. Techniques Avariety of techniques were used to compile the map of MiiIler Ice Shelf, depending upon the format of the data. The ice fronts identified on the oblique aerial photographs were sketched in by hand, those from satellite images digitized directly from the photographic product or digital data on screen, and vertical aerial photographs were interpreted using a radial-line plotter. Once the ice fronts had been drawn onto an existing base map, the cartographic data were edited using a Laser-Scan software package for editing geographic information and map data. The initial base map had been compiled by conventional cartographic methods, at 1:250 000 scale, for a BAS geological project (Moyes et al. 1994) and digitized subsequently. The details of the base map were improved by overlaying it on the 1986 digital Landsat image and, with the help of the IfAG photographs, the coastline and grounding line were digitized at a larger scale. Oblique aerialphotographs. By far the most approximate ice fronts mapped were those sketched from the oblique aerial photographs. These were estimated by noting significant features on the photographs (e.g. the position of the ice front at the mainland coast and at Humphreys Ice Rise) and extrapolating a line from these points. Satellite images. The satellite images, available as photographic products at 1:250 000 scale and as digital data, are of a resolution thh; clearly depicts the ice front. Using known control points from the BAS triangulation network, the images could be positioned accurately with respect to the base map. Ice fronts were digitized directly using a graphics package, which allowed accurate mapping to better than 100 m. Vertical aerial photographs. Radial distortion from the principle point of the image is enhanced by the change in elevation of the ground surface, although on an ice shelf this change is negligible. Because large-scale photographs, such as the 1956 FIDASE series, would have produced a noticeable radial error in the ice front data, tracing was regarded as too inaccurate for this project. A Watts Radial-line plotter, which is designed to transfer planimetric detail directly from vertical stereoscopic photographs on to abasemap,was used instead. By using the radial-line triangulation, the displacement produced by relief and radial-lens distortion is compensated for by the intersectionofimages fromanoverlapping pair. The plotterwas used to createalarger-scalebasemap ofthecoast, groundingline and rock outcrop polygons of the area around Miiller Ice Shelf; these details were subsequently digitized and merged with the existingmap data to enhance the earlier map. Aftertransferring the digital data into a GIS package (ARCDNFO software) the Table I. Details of data sources used. Source Format Date Scale of data RARE Oblique aerial 1947 - photography FIDASE Vertical aerial 1956 1 2 7 000 photography TMA Oblique aerial 1963 - photography LANDSAT Satellite image 1974 1:250 000 (photographic product) (digital) photography photography LANDSAT Satellite image 1986 1:250 000 IfAG Vertical aerial 1989 1:70 000 BAS Vertical aerial 1993 1:25 000 1:20 000 BAS Digital topographic 1994 1:250 000 (Moyes et al.) base map RARE = Rome Antarctic Research Expedition, FIDASE = Falkland Islands Dependencies Aerial Survey, TMA = Trimetrogon aerial photographs, IfAG = Institut f u r Angewandte Geodasie, BAS = British Antarctic Survey. 197 https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0954102095000265 Downloaded from https://www.cambridge.org/core. Carnegie Mellon University, on 06 Apr 2021 at 01:16:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms https://doi.org/10.1017/S0954102095000265 https://www.cambridge.org/core 198 C.G. WARD Fig. 1. Map showing the retreat and advance of the Muller Ice Front between 1947 and 1993. The graph indicates changes in the area of Miiller Ice Shelf with time; patterned lines (with dates) correspond to ice- front ornaments on the map. areaoftheiceshelfcouldbecalculatedfor eachyearmapped and the variations with time assessed (Fig. 1). Results The map of the Miiller Ice Shelf (Fig. 1) with seven ice fronts indicates that the ice shelf has decreased in size over the 46-year period. However, the regression pattern is not simple since a rapid advance of the ice front occured between 1947 and 1956, increasing the ice shelf area from approximately 51 km2 to c. 78 kmz. A second period of ice front advance, 1974-1986, representsan area increase of c. 4 km2. These results do not agreewiththose documented by Domacketal. 1995, who found no noticeable change during the period 1947-1974 and from 1974 to the present day. Despitetherangeinquality ofthesource data, the variations in the position of the ice front with time, as recorded by this study, are considered to be real. The ice shelf does appear to be undergoing some form of change reaching a minimum extent in 1993. Over the past 46 years it has experiencedavariation ofc. 31 km2, 40%ofitsmaximumextent in 1956. Similar changes in ice shelf areas have been recorded elsewhere in the Antarctic Peninsula (Doake & Vaughan 1991, Skvarca 1993)) possibly reflecting recent climate variability in the region (Morrison 1990, King 1994). Acknowledgements I wish to thank Paul Cooper (British Antarctic Survey) for his help in providing the area analysis data, and the graph included in Fig. 1. Also Adrian Fox, Andrew Perkins and Janet Thomson (British Antarctic Survey) for their technical advice during this project. Theprojectisbeingcarriedout aspart of anundergraduate final year thesis. References DOAKE, C.S.M. & VAUGHAN, D.G. 1991. Rapid disintegration of the Wordie Ice Shelf in response to atmospheric warming. Nature, 350,328-330. DOMACKE.W.,ISHMAN,S.E., S ~ , A . B . , M C C E N W V C . E . &Jvu,A.J.T. 1995. Late Holocene advance of the Miiller Ice Shelf, Antarctic Peninsula: Sedimentologic, geochemical and palaeontologic evidence. AntarcticScience, 7,159-179. KING, J.C. 1994. Recent climate variability in the vicinity of the Antarctic Peninsula. International Journal of Climatology. 357-369. MORRISON, S.J. 1990. Warmest year on record in the Antarctic Peninsula? Weather, 45,231-232. M o n s , A.B., Wnuw, C.F.H., THOMSON,J.W. EIAL. 1994. Geologicalmapof Adelaide Island to Foyn Coast, BAS GEOMAP Series, Sheet 3,1:250 OOO scale map with supplementary text, 60 pp. Cambridge: British Antarctic Survey. SKVARCA, P. 1993. Fast recession of the northern Larsen Ice Shelf monitored by space images. Annals of Glaciology, 17,317-321. https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0954102095000265 Downloaded from https://www.cambridge.org/core. Carnegie Mellon University, on 06 Apr 2021 at 01:16:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms https://doi.org/10.1017/S0954102095000265 https://www.cambridge.org/core