A Gis‐Based Reconstruction of Little Ice Age Glacier Maximum Extensions for South Tyrol, Italy
Computers and Technology
Submitted By c716563
A GIS‐based reconstruction of Little Ice Age glacier maximum extensions for South Tyrol, Italy 5
Department of Geography University of Innsbruck
Department of Geography University of Innsbruck
Department of Geography Institute for Applied Remote Sensing University of Innsbruck EURAC Bolzano Keywords: Little Ice Age, glacier reconstruction, glacier development, GIS Abstract A reconstruction method of historical glacier topographies and a possibility of the usage of these results are demonstrated in this paper. This reconstruction was accomplished for 310 Alpine glaciers in South Tyrol, Italy. These glaciers are featured with a wealth of different historical (e.g. paintings, photographs and historical maps) and recent data sources (airborne laser scan based digital terrain model and digital orthophotos) that allow the reconstruction of the Little Ice Age maximum extension. These sources are among the best historical and recent documents of glaciers for the mid 19th century. The results of this reconstruction visualize the ongoing climate change in a comprehensive way. The area changes between the time of the Little Ice Age maximum extent (around the year 1850) and the recent glaciation in 2006 amounts in a loss of 182.4 km² or almost 66 %. In the same time the calculated mean equilibrium line altitude for all South Tyrolean glaciers rose approximately by 160 m. Address for correspondence: C. Knoll, Department of Geography, University of Innsbruck, Innrain 52, A ‐ 6020 Innsbruck, Austria. E‐mail: firstname.lastname@example.org. 1 Introduction The worldwide downwasting of glaciers since the end of the Little Ice Age (LIA) is one of the most significant facts demonstrating the impact of climate change [IPCC 1995; IPCC 2007]. It shows that glacier changes as a result of global climate change are easy to observe. During the last phase of the Little Ice Age in the mid 19th century the majority of the world's glaciers reached maximum extensions. Among other mountain groups, in the European Alps this can be derived from historical maps, dendrochronology, mapping of the conspicuous glacier forefields and moraines and a wealth of historical evidence [Maisch et al. 1999; Zumbühl and Holzhauser 1988]. In general, available historical records can give a detailed picture of former glacier extents and the fluctuations [Gross 1987; Grove 2004; Grove 2004; Maisch, Wipf, Denneler, Battaglia and Benz 1999; Patzelt 1980]. 1
The worldwide first glacier inventory has been compiled for the Eastern Alps by Richter [Richter 1888] in the year 1888 based on the 3rd Austro‐Hungarian topographic survey. Apart from many local studies, regional or national inventories in the European Alps such inventories have been compiled within the last decade i.e. by Patzelt and Gross [Gross 1987; Patzelt 1978; Patzelt 1980], Maisch [Maisch et al. 1993; Maisch, Wipf, Denneler, Battaglia and Benz 1999], Paul [Paul 2004; Paul and Andreassen 2007; Paul et al. 2002] and Zemp [Zemp et al. 2007; Zemp et al. 2004]. In Austria, LIA reconstructions have been made by Patzelt [Patzelt 1967; Patzelt 1973] and Gross [Gross 1987]. Of the investigation area LIA reconstructions exist only for two parts, one for the Hinteres Martell Valley in the Ortler – Cevedale Group [Müller 2006] and one for the Rieserferner Group [Damm 1998]. In this paper, we present the methods and results of the reconstruction of the LIA maximum glacier extent for South Tyrol (Italy). It has been recognized that GIS is an important and useful tool for this study providing the means to capture and store information of the glacier extents of the region of South Tyrol and visualize it with "Visual Nature Studio" for further presentations. The spatial glacier extent of the Little Ice Age maximum can be reconstructed with the help of raster data calculated from Airborne Laser Scanning (ALS) point data from 2004/2005, digital orthophotos from 2006, historical maps of the 3rd and 4th Austro‐Hungarian topographic survey ("Franzisco‐Josephinische Landesaufnahme" and "Präzisionsaufnahme") and from historical paintings by Thomas Ender [Ender 1964]. We tried to integrate these data into a geodatabase. Unfortunately some of the historical map data are difficult to access or georeference with the required accuracy. The most reliable archive data source is the 4th topographic survey of the Habsburg Empire which was surveyed from 1896 until 1914. The georeferenced sheets of this survey serve as the principal data source for the detection of vanished glaciers and of glacier extents around the turn of the century. The maps of the 3rd topographic survey of the Austro‐Hungarian Empire were surveyed between 1869 and 1887. During both mapping campaigns, the individual map sheets of South Tyrol were created within a decade. Hence the glacier extent depicted on the maps can be assumed to be contemporaneous. The ALS digital elevation model (DEM) and the orthophotos are of highest resolution and accuracy standards. 2 The Study Area Our study focuses on the glacierized areas of the Italian Autonomous Province of Bozen/Bolzano – South Tyrol. It is located in the north east of Italy and is the border province to Austria and Switzerland. The mountains of South Tyrol are situated south of the alpine main divide. The principle mountain groups are the southern Zillertal Alps, the Dreiherren Group, the Rieserferner Group, the Dolomites, the Ortler – Cevedale Group, Sesvenna Group, the Ötztal Alps, the Texel Group and the Stubai Alps (Figure 1). All of these ranges drain into the river Etsch and subsequently Adriatic Sea. The glaciers along the alpine main ridge (Sesvenna Group, Ötztal Alps, Texel Group, Stubai Alps and Zillertal Alps) are primarily exposed towards the south‐east and south‐west, while the glaciers farther to the south are mostly facing towards the north‐east, north and north‐west.
Figure 1 Little Ice Age maximum glaciation in South Tyrol, Italy.
3 Used data sources for the reconstruction 3.1 Digital Elevation Model The ALS DEM used in this study is a 2.5 m raster dataset of South Tyrol (see Fig. 2) which is calculated from ALS point data [Wack and Stelzl 2005] with a nearest neighbour interpolation method on the last pulse returns. The initial ALS point dataset was acquired using the "Optech" and the "Terrasys" scanner systems during late summer 2005. The elevation accuracy is 0.4 m below 2000 m a.s.l. and 0.55 m above 2000 m a.s.l. An average point density of 8 points per 25 square meters for areas below 2000 m a.s.l. and 3 points per 25 square meters for areas above 2000 m a.s.l. was achieved. The full information of the collected ALS point data, i.e. information for first pulse, last pulse and intensity, is stored in a separate dataset that has not been released yet.
Langtauferer Ferner, Ötztal Alps. Mapping year 2005.
Figure 2 Clip of Hillshade of the ALS DEM, sheets 12041‐4, 12051‐4, 13041‐4 and 13051‐4, showing the area of
3.2 Digital Orthophotos Survey flights to obtain digital aerial photographs of the whole area of South Tyrol (Fig. 1) were carried out by the company Compagnia Generale Ripreseaeree S.p.A. (Parma, Italy) during summer 2006. The orthophotos (see Fig. 3) are calculated at a scale of 1:10.000; all photographs were taken in natural colour with a CCD‐digital camera. The orthophotos are calculated from scanning stripes with a planimetric resolution specified as 1 pixel (pixel size equal to 0.5 m). The planimetric accuracy as stated by the company is equal to ±2 m. Basic information for the photogrammetric data acquisitions have been recorded during the scanning flight with an integrated GPS. Using the orthophotos as a base layer in a GIS environment we delineated the LIA glacier boundaries as polygons by visual inspection.
Figure 3 Clip of orthophoto 1:10.000, sheets 12040, 12050, 13040 and 13050, showing the area of Langtauferer Ferner, Ötztal Alps. Mapping year 2006.
3.3 Historical Topographic Maps For this study we used two different historical map datasets. The first one is the 3rd topographic survey and the second is the 4th topographic survey of the Austro‐Hungarian Empire. Both survey campaigns were designed with the rapid production of maps at a scale of 1:75.000 for the whole Empire in mind. Between the years of 1869 and 1887 the Austro ‐ Hungarian army's “k. u. k. Militärgeographisches Institut” carried out the field survey for the 3rd topographic survey (see Fig. 4). This so‐called "Franzisco‐Josephinische Landesaufnahme" was recorded with the plane‐ table method and drawn at a scale of 1:25.000 using the Joseph Marx Baron of Lichtenstern polyhedron projection. The individual map sheets are hand coloured drawings ("Originalaufnahme"). Altitudes were determined by triangulation and barometers. For various reasons, the content of the maps is occasionally rather poor and inaccurate. This includes especially the warpage of several map sheets, position errors and occasionally the oversimplification of the topography of mountain ranges. The lack of accuracy was mainly due to varying survey methods for the map sheets, lack of experience of the surveying teams in high mountain areas, weather conditions, snow cover and logistical problems, which occasionally 4
caused a misinterpretation of the topography [Hofstätter 1989; Hofstätter 1989]. As a consequence, georeferencing was considered totally not acceptable. We tried a re‐ georeferencing of the single map sheets with the radial basis function and the piecewise affine method. The results were good but they could not compensate the missing geographic contents. Hence the maps of the 3rd topographic survey were only used to locate glaciers that do not exist anymore.
Figure 4 Clip of the "Sektionsblatt" 1:25.000 of the 3 topographic survey, sheet 5245/4, showing the area of Langtauferer Ferner, Ötztal Alps. Mapping year 1872. (© BEV 2008, T2008/50216). rd 145
Langtauferer Ferner, Ötztal Alps. Mapping year 1910. (© BEV 2008, T2008/50216).
Figure 5 Clip of the "Sektionsblatt" 1:25.000 of the 4th topographic survey, sheet 5245/4 showing the area of
In anticipation of hostilities with the Kingdom of Italy, the 4th topographic survey (see Fig. 5) was carried out by the Austro ‐ Hungarian army's “k. u. k. Militärgeographisches Institut” from 1896 to 1914. This so‐called "Präzisionsaufnahme" was also drawn and coloured by hand at a scale of 1:25.000 using the Joseph Marx Baron of Lichtenstern polyhedron projection. However, in consideration of the weaknesses of the 3rd topographic survey, terrestrial stereo‐ photogrammetry was used as the state‐of‐the‐art method. The geodetic base was the MGI (Militärgeographisches Institut) Ferro datum based on the Bessel 1841 ellipsoid and the Gauss‐ Krueger projection. This historical map set is characterized by higher spatial resolution and more details in the map contents with a more accurate georeferencing than the older map set of the 3rd topographic survey. Information about applied methods, mapping rules and used definitions are summarized by Hofstätter (1989a, 1989b). Digital copies of the hand‐drawn original sheets 1:25.000 were used for this work. Compared to modern practices of map survey and production, the accuracy of the maps is supposed to be lower but was at the highest level at the time of recording and is the best historical data that is accessible for the whole South Tyrolean area. 4 Methods Reconstructing the Little Ice Age maximum extent of glaciers requires detailed geomorphologic mapping via high resolution DEMs, orthophotos and field work. This requires sufficient geomorphologic evidence, usually lateral and terminal moraines and trimlines, to allow the reconstruction of the former glacier topography. In the case of the LIA maximum moraines in the investigation area, field evidence and documentary evidence in the ablation areas is so conspicuous that misinterpretations of landforms by a trained glacial‐geomorphologist who is familiar with the examination area can be practically excluded. In contrast to the ablation areas the reconstruction in the accumulation areas is more difficult because there almost no moraines occur and the reconstruction is only based on the experience of the surveyor and the calculated heights [Maisch, Wipf, Denneler, Battaglia and Benz 1999; Patzelt 1967; Patzelt 1973]. Careful consideration of the field evidence is critical for evaluating the timing and extent of glacier advances. The exact year of the maximum extent is only known for a few large valley glaciers like Sulden Ferner and Langtauferer Ferner. However, the temporal uncertainty of the maximum is only within a range of a few years. For the reconstruction of the Little Ice Age glacier topography we used the interpretation of the new ALS DEM and orthophotos combined with field inspection of selected sites to reconstruct area and volume changes and the equilibrium line altitude (ELA). The calculation and reconstruction is based on the interpretation of the 3rd and 4th Austro‐ Hungarian topographic survey to find glaciers that do not exist anymore as well as on the interpretation of the hillshade of the ALS DEM and the orthophotos. The transformation and visualization of the 2D data into 3D data with ESRI ArcScene and Visual Nature Studio gave additional useful insights into the maximum extent. Field work formed the basis for closing the gaps. For the contour line reconstruction a cartometric interpretation of the generated 100 m contour lines of the ALS DEM was combined with shear stress calculations of the glacier tongue. The shear stress τ was derived from a mean slope over two 100 m height intervals. In a first 6
approximation, the quantity of τ depends on the slope of the glacier bed. Then the ice thickness is [Paterson 2001] α for the contour line reconstruction, where τ is between 50 ‐ 300 kPa depending on the steepness of the glacier, ρ is the density of glacier ice [900 kgm‐³] and g is the acceleration due to gravity [9.81 ms‐2] [Paterson 2001]. The areal extent of the glaciers is described by closed polygons. In the case of the three digitized glacial stages (LIA maximum, 1997 and 2006) altitude data for each vertex of the polygons was interpolated from the respective DEM. In order to be consistent with the computed DEMs the photogrammetrically derived z‐values of the polygons of the 1997 survey were also replaced by the respective height values of the DEM. 5 Quantification of glacier retreat At the LIA maximum, 310 glaciers existed in South Tyrol. The three largest glaciers were at the time of maximum extension Übeltalferner (Stubai Alps, 12.41 km²), Zufall Ferner (Ortler– Cevedale Group, 14.88 km²) and Suldenferner (Ortler – Cevedale Group, 10.18 km²). 13 glaciers (4.2 per cent) were larger than 4 km² and 238 (76.8 per cent) out of the 310 were smaller than 1 km². The total area of the South Tyrolean glaciers was 276 km² (Table 1). It is now possible to determine the area and volume loss of glaciers since the LIA maximum at about 1850 for the first time for the entire South Tyrolean mountains. Table 1 gives an overview of the glacierized area at the time of maximum glaciation during LIA and the recent glacier area of 2006 [Knoll and Kerschner 2009 (submitted)] in the respective mountain ranges. It shows that about two thirds of the glacier area was lost during the last approximately 150 years. The equilibrium line altitude was calculated for each glacier and a mean value for each of the mountain ranges (see Figure 1). The accumulation area method with an accumulation area ratio (AAR) of 0.67 is used for this purpose, i.e. the accumulation area is 67% of the total glacier surface, was used for the calculation of the ELA change since the time of maximum extension [Gross et al. 1978]. A steady‐state of the South Tyrolean glaciers was estimated for both times whereas a steady‐state in 2006 is not to be expected. The effective rise of the ELA was larger than the mean value of 160 m (Table 1), because in 2006 and the preceding years glaciers were clearly out of balance. A value of approximately 250 m, as it can be derived in the Zillertal Alps is most probably more realistic, because glaciers there are mainly situated in small cirques where ice reacts quickly to climatic changes. ρ h
Table 1: Area and ELA changes between LIA maximum glaciation and 2006.
6 Visualization of the glacier development since the end of Little Ice Age 235
6.1 Data preparation Based on digitised contour lines a digital elevation model with a resolution of 2.5 m has been created in ENVI 4.5 (Convert contours to DEM – linear technique). Subsequently, the difference to the more recently released ALS DEM 2006 of South Tyrol was calculated. The obtained result was a signed floating ‐ tif imagery that had to be separated in two different files, one with only positive integer values and another with only negative integer values. This is, because the visualization program Visual Nature Studio 2.8 is only able to read 8bit unsigned integers for its function “Area Terraffector”. 6.2 Data visualisation The ALS DEM of South Tyrol and the digital orthophotos were imported into the program Visual Nature Studio 2.8 for data visualisation. Additionally, the difference models (both the files positive and negative values) were imported and connected to the “Area Terraffector” function to modify the terrain. As a second last step the white colour was allocated to the glacier body in order to achieve a more realistic impression. Finally, a historic image as well as a recent image from the same position was generated to allow direct comparison.
(right, 2006) extension.
Figure 6 and 7: Visualisation of Langtaufererferner, Ötztal Alps/Italy for its maximum (left, about 1830) and recent
7 Discussion and Conclusions Recent and historic remote sensing technologies combined with suitable reconstruction techniques can contribute to a reconstruction of LIA maximum glacier extents and recent Alpine glaciers topographies. This study is focused on the historic glaciers where a realistic reconstruction of LIA glaciers and the corresponding contour lines was achieved. In the beginning we tried a reconstruction based on historical map sets but, because of problems in georeferencing and topographic contents, these maps of the 3rd and 4th Austro‐Hungarian topographic survey have not been taken into account. In the next step the work was focused on the reconstruction of the LIA maximum extents based on the interpretation of digital orthophotos of 2006 and the ALS DEM of 2005. A detection of lateral and terminal moraines would be half‐automatically possible but will give partly incorrect results as moraines are often disconnected or have been eroded during approximately the last 150 years. Also a manual detection is only possible if the adapter is a trained glacial‐ geomorphologist who is familiar with the examination. A manual detection method for the moraines was chosen for the present LIA reconstruction. This new approach, which is based on ALS DEMs and orthophotos, provides a geometrical monitoring and verification of the reconstructed results. Former reconstructions have been based mostly on historical or older topographical maps where the analysis of the historic glaciological topographies was drawn with ink or pencil on transparent paper. The analysis of these results was accomplished with mechanical planimeters. Today comprehensive topographic fundamentals allow more precise investigations. Glacier topographies can be reconstructed easier with much higher precision and accuracy because of higher resolutions of the input data. The main advantage of this approach is the doubtless, well‐defined and efficient operation method providing a good basis for further analysis of the changed climatic conditions since the end of LIA. Acknowledgements The project is funded by the PhD scholarship of the University of Innsbruck and by the Faculty for Geo‐ and Atmospheric Sciences, University of Innsbruck. The Authors want to thank Gernot Patzelt, Georg Kaser, Michael Kuhn, Astrid Lambrecht, Jakob Abermann and the Glaciological Seminar Innsbruck (MSc, PhD and PostDocs) of the University of Innsbruck and the Austrian topographic survey (Bundesamt für Eich‐ und Vermessungswesen BEV) for the maps of the 3rd and 4th topographic survey (© BEV 2008, T2008/50216). References DAMM, B. 1998. Der Ablauf des Gletscherrückzuges in der Rieserfernergruppe (Tirol) im Anschluss an den Hochstand um 1850. Zeitschrift für Gletscherkunde und Glazialgeologie 34, 141‐159. ENDER, T. 1964. Katalog zur Ausstellung "Thomas Ender" der graphischen Ausstellung der Albertina. Albertina Vienna, Vienna. GROSS, G. 1987. Der Flächenverlust der Gletscher in Österreich 1850–1920–1969. Zeitschrift für Gletscherkunde und Glazialgeologie 23(2), 131‐141. 9
GROSS, G., KERSCHNER, H. and PATZELT, G. 1978. Methodische Untersuchungen über die Schneegrenze in alpinen Gletschergebieten. Zeitschrift für Gletscherkunde und Glazialgeologie 12(2), 223‐251. GROVE, J.M. 2004. Little ice ages : ancient and modern. Routledge, New York. GROVE, J.M. 2004. Little ice ages : ancient and modern. Routledge, New York. HOFSTÄTTER, E. 1989. Beiträge zur Geschichte der Österreichischen Landesaufnahmen: Ein Überblick der topographischen Aufnahmeverfahren, deren Ursprünge, ihrer Entwicklungen und Organisationsformen der vier österreichischen Landesaufnahmen. Bundesamt für Eich‐ und Vermessungswesen BEV, Vienna. HOFSTÄTTER, E. 1989. Beiträge zur Geschichte der Österreichischen Landesaufnahmen: Ein Überblick der topographischen Aufnahmeverfahren, deren Ursprünge, ihrer Entwicklungen und Organisationsformen der vier österreichischen Landesaufnahmen. Bundesamt für Eich‐ und Vermessungswesen BEV, Vienna. IPCC 1995. Climate Change 1995: The Science of Climate Change. Contribution of Working Group I to the Second Assessment of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. IPCC 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, S. SOLOMON, D. QIN, M. MANNING, Z. CHEN, M.C. MARQUIS, K. AVERYT, M. TIGNOR and H.L. MILLER Eds. Intergovernmental Panel on Climate Change, Cambridge, New York. KNOLL, C. and KERSCHNER, H. 2009 (submitted). A glacier inventory for South Tyrol, Italy, based on airborne laser scanner data. Annals of Glaciology 53. MAISCH, M., BURGA, C.A. and FITZE, P. 1993. Lebendiges Gletschervorfeld : von schwindenden Eisströmen, schuttreichen Moränenwällen und wagemutigen Pionierpflanzen im Vorfeld des Morteratschgletschers Zurich. MAISCH, M., WIPF, A., DENNELER, B., BATTAGLIA, J. and BENZ, C. 1999. Die Gletscher der Schweizer Alpen. Gletscherhochstand 1850, Aktuelle Vergletscherung, Gletscherschwund Szenarien, Schlussbericht NFP31. VdF Hochschulverlag, Zurich. MÜLLER, S. 2006. Gletscherstände und Klimawandel im Hinteren Martelltal, Südtirol. In Fakultät für Forst‐ und Umweltwissenschaften University of Freiburg, Freiburg im Breisgau. PATERSON, W.S.B. 2001. The Physics of Glaciers. New York. PATZELT, G. 1967. Die Gletscher der Venedigergruppe ‐ Die Geschichte der Schwankungen seit Beginn der postglazialen Wärmezeit. In Department of Geography University of Innsbruck, Innsbruck. PATZELT, G. 1973. Die neuzeitlichen Gletscherschwankungen in der Venedigergruppe (Hohe Tauern, Ostalpen). Zeitschrift für Gletscherkunde und Glazialgeologie 9, 5‐57. PATZELT, G. 1978. Der Österreichische Gletscherkataster. In Almanach ’78 der Österreichischen Forschung, Vienna, 129‐133. PATZELT, G. 1980. The Austrian glacier inventory: status and first results. IAHS. PAUL, F. 2004. The new Swiss glacier inventory 2000 ‐ Application of remote sensing and GIS. In Department of Geogrpahy University of Zurich, Zurich, 194.
PAUL, F. and ANDREASSEN, L.M. 2007. A new glacier inventory for the Svartisen area (Norway) from Landsat ETM+: Methodological challenges and first results. In Proceedings of the Workshop and GLACIODYN (IPY) meeting, Pontresina, 15‐18 January 2007 2007. PAUL, F., KÄÄB, A., MAISCH, M., KELLENBERGER, T.W. and HÄBERLI, W. 2002. The new remote‐ sensing‐derived Swiss Glacier Inventory: I. methods. Annals of Glaciology 34, 355‐361. RICHTER, E. 1888. Die Gletscher der Ostalpen. Engelhorn, Stuttgart. WACK, R. and STELZL, H. 2005. Laser DTM generation for South‐Tyrol and 3D‐Visualization. In Proceedings of the ISPRS Workshop Laser scanning 2005, G. VOSSELMAN and C. BRENNER Eds., Enschede, the Netherlands, 48‐53. ZEMP, M., PAUL, F., HOELZLE, M. and HÄBERLI, W. 2007. Glacier fluctuations in the European Alps 1850–2000: an overview and spatio‐temporal analysis of available data. In The darkening peaks: Glacial retreat in scientific and social context, B. ORLOVE, E. WIEGANDT and B. LUCKMANN Eds. University of California Press. ZEMP, M., PAUL, F., HOELZLE, M. and HAEBERLI, W. 2004. Alpine glacier fluctuations 1850‐ 2000: overview and spatio‐temporal analysis of available data and its representativity. In International and Interdisciplinary Workshop on Mountain Glaciers and Society, October 6‐8, 2004., Wengen, Switzerland. ZUMBÜHL, H.J. and HOLZHAUSER, H. 1988. Alpengletscher in der Kleinen Eiszeit. In Die Alpen. Zeitschrift des Schweizer Alpen‐Club SAC, 129‐322.