Remapping of the Cathedral Massif Glacier (B.C., Canada) - from Traditional Mapping to Digital Techniques Walter Gruber walter.gruber@sbg.ac.at Heinz Slupetzky Heinz.Slupetzky@sbg.ac.at Department for Geography and Geoinformation University of Salzburg, Austria Abstract Part I (Heinz Slupetzky) Surveying of the Cathedral Glacier for glaciological purposes within the Juneau Icefield Research Project JIRP 1972 - 1985 Climatic changes and trends are clearly reflected in the reaction and change of glaciers. Long time series of detailed mass balance measurements give the understanding of climate/glacier relationship. To record glacier fluctuations, especially in their long-term variability, it is necessary to gain various informations. One of the most important are large scale maps. F. e. for mass balance studies and to record the three dimensional change of glaciers a repeated survey is needed. While three, four decades ago classical photogrammetric methods were used specific applications and improvements from satellite techniques to digital ones enable one nowadays to get informations of glaciated areas all over the world and not only from a limited numbers of glaciers (Haeberli, 1998; Roethlisberger, 1986). But still large scale maps are needed. In remote areas mapping has still an expeditionary character like the mountaneous area which the present paper is dealing with. Since 1946 a long-term research Program, the Juneau Icefield Research Program (JIRP), was established to investigate several glaciers in the area between Alaska and British Columbia. This program is organized and coordinated by the Foundation for Glacier and Environmental Research, directed by M.M. Miller since then, later in cooperation with the University of Idaho, Moscow ID. Surveying and mapping of glaciers are part of the interdisciplinary research within the JIRP (Welsch, W., M. Lang & M.M. Miller, Ed., 1997). The Juneau Icefield is situated in the northern Boundary Range, the northernmost range of the Coast Mountains in British Columbia and covers about 3000 km2 of interconnected highland glaciers, composing the fifth largest icefield in the western hemisphere. The Cathedral Massif The Cathedral Massif is located at the northeastern periphery of the main Icefield, near the south end of Atlin Lake in northwest British Columbia. A small cirque glacier (approx. 1.6 km2) north of the highest Peak (Cathedral Peak, 2315 m) called Cathedral Massif Glacier, is part of the areas which are investigated within the JIRP. In 1972 a camp (Camp 29) for the research activities was established. Since that time several scientists have done contributions to the geomorphology, neoglacial chronology, climate and glaciology of this area (Jones, 1975). The mapping of the Cathedral Massif Glacier 1972-1985 For the investigations an appropriate basemap was necessary therefore a first reconnaissance was done in 1972 by G. Konecny. In 1975, geodetic field surveys as well as terrestrial photogrammetry were carried out by a team of surveyors of the Technical University of Hannover, Germany and one year later a two-colour map 1:5000 was produced and published. (Konecny et al., 1976). One of the purposes was monitoring of a glacier on the dry, continental side of the Juneau Icefield in comparison to the Lemon Creek Glacier, Alaska, on the wet maritime side ( AGS, 1960); remapping has been done there in 1989 (MARCUS, G. et al.1997 pp.185). Application of the map for mass balance purposes The “Konecny-map” was used by H. Slupetzky to make a field map for mass balance studies. The mass balances of 1976/77 and 1977/78 were measured during his stay at the University of Idaho, Moscow, ID. 1976/77. At the first time not only the yearly balance was measured but also temporary balances were measured weekly. The computation was done by R. Hasenauer based on the data collected by H. Slupetzky. The Cathedral Massif Glacier had a negative balance in both years: Specific net mass balance 76/77 - 39 cm and 77/78 - 69 cm and and a net mass balance of - 632 000 m_ and -1 106 000 m_ resp. (Hasenauer, R. 1984). Within the framework of mass balance studies on the glacier fieldwork for a new map was done by H. Slupetzky, L. Mauelshagen and F. Boochs. The geodetic computations and photogrammetric compilation using terrestrial and aerial photos was done by L. Mauelshagen, F. Boochs and W. Schröter at the University of Bonn, Germany. H. Slupetzky carried out the compilation, geomorphological interpretation and the drafting of the map, cartographic work was done by W. Gruber. This 5-coloured map published 1985 covered both the glacier and the forefield, and was made using traditional manual techniques for drafting and reproduction (Slupetzky H., L. Mauelshagen & W. Gruber, 1988). One of the main goals to remap the Cathedral Massiv Glacier in 1999 was to estimate the volume change since the first survey(s) and another to correct the first map(s). The maps of 1976 and 1985 are both based only on a local geodetic network - which was established by KONECNY et al. in 1975 - due to the lack of connection with the official Canadian UTM-Net. Elevations are based on barometric altimetry showing a difference of close to 200 m compared to official maps 1: 25000 an 1:50000 which were available later. Part II (Walter Gruber) Remapping of the Cathedral Massif Glacier 1999 by the means of digital techniques – methods and perspecvtives a) Fieldwork Due to the deficits of the local geodetic framework and the altimetry, a new surveying of the glacier was planned and carried out in 1999. Since the last geodetic fieldworks in 1977 the GPS (Global Positioning System) has replaced traditional techniques for surveying of remote areas. GPS has also been used for the geodetic activities on the Juneau Icefield since 1992. In August 1999 a team of geographers from the University of Salzburg, headed by H. Slupetzky, carried out the geodetic surveying of Cathedral Massif Glacier using Trimble Pathfinder GPS-Receivers (Trimble Inc, 1996). For the transformations of old coordinates several points on the glacier and the vicinity were collected. The actual glacier margin was stored as a line of kinematic points. From 13 stations 78 terrestrial photos were taken by a calibrated medium-format camera (Mamiya RB 67) for photogrammetric compilation. Additional observations and investigations concerning the topography and chronology of the glacier and the forefield were also carried out. b) Compilation of field-data Processing of GPS-Data: Computation of the GPS-data was courtesy provided by M. Welsch and M. Lang from the University of the German Armed Forces in Munich-Neubiberg, Germany. The data were computed using GPS-observations from surveying activities on the Juneau Main Icefield and from the control stations at Whitehorse and Juneau. The final coordinates are accurate within 0.35 m (horizontal) and 0.25 m (vertical) giving a very good source for transformations and controlling the photogrammetric compilations (Lang, M., 2001). Photogrammetric Plotting: For the now digital compilation of the photogrammetric images the original 6x7 cm –slides were scanned at a resolution of 1200dpi by a HP-Scanner. Photomodeller Pro was used for compiling these digital TIFF-images. This software was primarily designed for low-range photogrammetry but as the area of the Cathedral Massif Glacier is small (2 x 2 km) the compilation should be successful. The program needs three or more images, taken from different stations, which are overlapping the area to be compiled. On the screen identical points must be marked on the different images and then the software is able to compute 3d-coordinates from the rays of the different camera stations (Eos-systems, 1997). Using known control points the model-coordinates can be transformed into a user-defined coordinate system. For the new model of the Cathedral Massif Glacier 39 Images from 13 stations and several photos taken from a helicopter were used to extract more than 300 points covering the topography and some control points for the transformation. The final coordinates have an accuracy from about 5-6 m RMS. Due to the high contrast of some images it was difficult to identify several points in the shadow zone of the southern/southeastern ridgeline, another problem was identifying small objects (rocks) on the surface of the glacier. c) Digitising and transforming the old maps There were no digital data from the maps of 1976 and 1985 so the prints were scanned and digitised manually on the screen. This task was done using the CAD-Software AutocadMap. To increase the accuracy the original foils were used for digitising. After that the digitised map-coordinates were transformed using ArcCad14 showing a sufficient accuracy. e) Other sources: the TRIM-Model The Surveying Authority of British Columbia has compiled digital terrain data for the whole province for many years. These data are based on aerial photos and include elevation points, breaklines, and topographic lines compiled in a work scale of 1:20000. For the Cathedral Massif Glacier Landdata BC provided the tile number 104m40 in the native SAIF-format, which was translated to 3d-DXF using the software FMBC. Points and lines of the TRIM-Model were imported into AutocadMap to overlay it with the coordinates of the scanned maps and the GPS-data. f) Computing the Digital Terrain Model (DTM) The DTM was computed from coordinates of the photogrammetric model exported from AutocadMap. The Kriging-Algorithm of the program Surfer version 7 (Keckler, D. 1997) was used to calculate the DTM. To cover the areas outside the photogrammetric model, the coordinate of the TRIM-Model were added. To compare the new model with the situation of 1977 another DTM was generated from coordinates of the digitised map. DTM are important tools for the analysis of topographic areas and the surface of glaciers as well as a basis for the production of orthophotos. g) Visualization Digital Terrain Models and 3d-lines are a good background for several methods of topographic analysis as well as for a wide range of cartographic and multimedia representation. There are a lot of DTM-programs (like Surfer) and GIS-software which are able to visualize digital 3d-data on a computer-screen. Visualizing data on a computer is very flexible but in many cases a traditional printed map is also necessary and therefore is still an important cartographic product. As the screen of a computer is flat, there are some geometric methods to give the user a 3d-impression but there are still some approaches on “real” 3d-visualization. Most types of 3d-visualization are based on DTMs. There are many aspects of cartographic visualization which cannot be discussed here but: Contour Levels, Wireframe-Models, Shaded Relief and Profiles are some types of visualization which were applied to the new DTM of the Cathedral Massif Glacier. Conclusion The transition from traditional manual map production to modern digital techniques should be demonstrated on the new map of the Cathedral Massif Glacier (B.C., Canada). GPS-Surveying combined with digital photogrammetry allows to produce a Digital Terrain Model on a Desktop-PC. DTMs are very flexible for analysing and visualizing terrains. A lot of different software is able to apply digital techniques but if you must digitise analogue materials this can be very tedious. The transformation and combining of different sources can also be a problem. For a simple photogrammetric plotting or a standard GPS-survey there is no need for a specialist and for reasonable visualization a fully-trained cartographer is not always necessary. But enhanced knowledge of photogrammetry, geodesy and cartography plus appropriate software are a good background for producing a cartographic model. As in the case of the map 1:5000 from 1985 the new cartographic model of the Cathedral Massif Glacier is a result of applied geodesy, photogrammetry, cartography and glaciology but now the technology has changed from analogue to digital techniques.