mapping glaciers in jotunheimen, south-norway, during the ...elevation model. “little ice age”...

The Cryosphere 3 231ndash243 2009wwwthe-cryospherenet32312009copy Author(s) 2009 This work is distributed underthe Creative Commons Attribution 30 License

The Cryosphere

Mapping glaciers in Jotunheimen South-Norway during the ldquoLittleIce Agerdquo maximum

S Baumann1 S Winkler1 and L M Andreassen2

1Department of Geography Physical Geography University of Wuerzburg Wuerzburg Germany2Norwegian Water Resources and Energy Directorate (NVE) Oslo Norway

Received 16 June 2009 ndash Published in The Cryosphere Discuss 29 June 2009Revised 23 November 2009 ndash Accepted 24 November 2009 ndash Published 1 December 2009

Abstract The maximum glacier extent during the lsquoLittle IceAgerdquo (mid 18th century AD) in Jotunheimen southern Nor-way was mapped using remote sensing techniques Inter-pretation of existing glaciochronological studies analysis ofgeomorphological maps and own GPS-field measurementswere applied for validation of the mapping The length ofglacier centrelines and other inventory data were determinedusing a Geographical Information System (GIS) and a DigitalElevation Model ldquoLittle Ice Agerdquo maximum extent for a to-tal of 233 glaciers comprising an overall glacier area of about290 km2 was mapped Mean length of the centreline was cal-culated to 16 km Until AD 2003 the area and length shrankby 35 and 34 respectively compared with the maximumldquoLittle Ice Agerdquo extent

1 Introduction

Investigations of the glacier maximum extent during ldquoLit-tle Ice Agerdquo (LIA) in South Norway have until recentlymainly been carried out as locally focused studies on selectedglaciers (eg Faeliggri 1948 Hoel and Werenskiold 1962Matthews 1977 2005 Erikstad and Sollid 1986 Bogenet al 1989 Winkler 2001) These investigations includeddating of moraines eg by application of lichenometry andmapping of selected glaciers and glacier forelands Previ-ous studies have mainly focused on the region of Jostedals-breen and a number of individual glaciers in Jotunheimeneg Storbreen in Visdalen

Correspondence toS Baumann(sabinebaumannuni-wuerzburgde)

Glaciers offer a high potential to serve as key indicatorsfor climate change (IPCC 2007) In this context a detailedknowledge of glacial chronology during the later Holocene isimportant as it serves as opportunity to verify forecasts andsimulations of future glacier behaviour Furthermore and es-pecially in Norway knowledge of the behaviour of glaciersand their response to changes in climate has a practical mean-ing because 98 of the domestic electricity is produced us-ing hydropower and 15 of the exploited runoff is derivedfrom glacierized river basins (Andreassen et al 2005)

Mapping the LIA maximum glacier extent by conventionalfield work is time consuming It is therefore usually appliedto investigate selected glaciers rather than a large sample ofglaciers or a whole region By investigating larger regionshowever a robust regional trend can potentially be detectedto avoid ndash or at least minimise ndash misinterpretations of spe-cific behaviours of individual ldquokeyrdquo-glaciers as representa-tive To study a whole region with a large number of individ-ual glaciers many difficult to access on the ground remotesensing provides a practicable alternative Studies in otherregions have already demonstrated the potential of satelliteimagery as an efficient tool for mapping of maximum LIAextents of glaciers on a regional scale (Solomina et al 2004Csatho et al 2005 Paul and Kaab 2005 Wolken 2006Paul and Svoboda 2009)

The aim of this study was to reconstruct the glacier areaduring LIA maximum in Jotunheimen and to create a glacierinventory for the LIA maximum The data were used to an-alyze the glacier area change from LIA maximum until AD2003 and to detect any possible spatial variability in glacierbehaviour in this region

Published by Copernicus Publications on behalf of the European Geosciences Union

232 S Baumann et al Mapping glaciers in Jotunheimen South-NorwayFig 1

Fig 1 Location of the study area Jotunheimen (inset) and glaciers (gt001 km2) with flowlines during the LIA maximum Letter codesdenote STD=Styggedalsbreen BUK=Bukkeholsbreen GRJ=Grjotbreen MEM=Memurubreen (Inset map ESRI Templates)

2 Study area

The study area of Jotunheimen is located in central SouthNorway (615 N 83 E) (Fig 1) It has a high-alpinecharacter and the highest points of Norway are locatedhere (Galdhoslashpiggen 2469 m asl Glittertind 2464 m asl)The present glaciers (ie AD 2003) are mostly small in-dividual valley-type and cirque-type glaciers separated bysteep rock-walls and range from about 1300 to 2300 m asl(Andreassen et al 2008) Long-term recorded mass bal-ance along a West-East profile in southern Norway revealsa strong gradient in mass balance with decreasing massturnover towards the more continental drier interior (egOslashstrem et al 1988)

The study area lies in a transitional zone between maritimeand continental climate and represents the most continentalglacier area in southern Norway (Oslashstrem et al 1988)

3 Timing of the LIA maximum in Jotunheimen

In contrast to Jostedalsbreen to the West there are no his-torical documents or images of the glacier area of Jotun-heimen that would allow a precise dating of the LIA max-imum Oslashyen (1893) and Hoel and Werenskiold (1962) re-spectively report only a vague eyewitness story about themaximum extent of Storbreen but no clear evidence thatcould be used for a distinct timing of the LIA maximumTherefore existing reconstructions of the glacial chronol-ogy in Jotunheimen prior to the first scientific measurementsand historical photo-record around AD 1900 are primarilybased on lichenometry supported by14C-datings lake sedi-ment analysis mire and colluvial profiles etc (eg Matthews

1974 1975 1977 2005 Innes 1985 Erikstad and Sollid1986 Winkler 2001 Matthews and Dresser 2008) Thesestudies suggest that the timing of the culmination of the LIAin Jotunheimen falls roughly between AD 1750 and 1800In West and Central Jotunheimen the outermost moraines ofthe LIA often date from around AD 1750 whereas in EastJotunheimen the related moraines tend to be a few decadesyounger ie date from around AD 17801800 (Fig 2 Win-kler 2002 Matthews 2005) Owing to the regional ecolog-ical conditions (glacier forelands above the tree lines rel-atively stable moraine ridges limited press on lichens byvascular plants and mosses usually sufficient humidity forlichen growth siliceous rock types etc) lichenometry has acomparable high temporal resolution and gives reliable re-sults (Matthews 2005) The accuracy of lichenometry indating the LIA maximum in this region are assumed to bewithin plusmn20 years for the more recent sophisticated studiesusing regional lichen growth curves (eg Matthews 2005)

The AD 1750 LIA maximum in West and Central Jotun-heimen is interpreted analogously to the mid-18th centuryLIA maximum at Jostedalsbreen (Bickerton and Matthews1993 Nesje et al 2008a) The primary climatologic causefor this advance is considered to be increased winter precipi-tation (Nesje and Dahl 2003 Nesje et al 2008a b De Jonget al 2009 Nesje 2009) By contrast the second half ofthe 18th century was dominated by below-average summertemperatures (Winkler 2001 Nordli et al 2005) In this pe-riod the outlets of Jostedalsbreen retreated slowly from theiroutermost LIA positions without a major readvance proba-bly because the drop in summer temperatures was accompa-nied by low winter precipitation (Bickerton and Matthews1993) In Jotunheimen and especially at the most continental

The Cryosphere 3 231ndash243 2009 wwwthe-cryospherenet32312009

S Baumann et al Mapping glaciers in Jotunheimen South-Norway 233Fig 2

Fig 2 Available LIA glacier outlines from the geomorphological maps divided after source Moraine type and timing of LIA maximumand GPS points of moraine walls measured 2008 are marked Glacier areas in the 1980s are shown in grey (glacier outline 1980s StatensKartverk N50 LIA maximum outlines and timing Erikstad and Sollid 1986 Winkler 2001 Matthews 2005)

glaciers to the East however these climatic conditions werefavourable for glacier growth Therefore it seems likely thatan advance culminating around 1800 in East Jotunheimenoverrode all potential older LIA moraines In the more mar-itime western and central parts of Jotunheimen glaciers ad-vanced simultaneously and nearly reached the previous 1750maximum position but without overriding and destroyingthe pre-existing terminal moraine resulting in double-ridgedterminal moraines (with a measurable difference in lichensizes) The outer ridge is then interpreted to date from themid- and the inner ridge from the late 18th century (Win-kler 2002) Double-ridged terminal moraines have not beenobserved in East Jotunheimen (Winkler 2001)

We decided to ignore the described variations in the timingof LIA maximum extent of the glaciers in the study region forsimplicity reasons Our following calculations thus assumea uniform LIA maximum at AD 1750 However when in-terpreting the climatic causes of the LIA glacier behaviour itmust be emphasized that the timing of the maximum varieswith sim50 years in the region

4 Material and methods

41 Study material

To generate LIA glacier outlines different data sources wereused

ndash a satellite image Landsat 5 TM Path 199 Row 17 9August 2003 cell size 30 mtimes30 m cloud cover 0

with little seasonal snow remaining The Landsat im-age delivered by Norsk Satellittdataarkiv was alreadyorthorectified The orthorectification was tested with14 check points by the Norwegian Water Resources andEnergy Directorate (NVE) (Andreassen et al 2008)

ndash 26 aerial photos in 140 000 taken in 1966 1976 and1981 by Fjellanger Wideroslashe (Fig 3) All photos werecloud-free and taken at the end of the ablation season(AugustSeptember) In this study the difference in datewas unimportant because these photos were only usedfor the purpose of mapping of LIA glacier extent Allphotos were original blackwhite contact copies and hadto be orthorectified and georeferenced

ndash geomorphological maps containing LIA outlines of to-tally 25 glaciers partially including detailed moraineages (Erikstad and Sollid 1986 Winkler 2001Matthews 2005) (Fig 2) These maps have no coor-dinate system so some have to be classified as sketchesrather as decent maps

ndash GPS points for LIA moraine ridges at eight glaciersThis data was collected AD 2008 during field-work us-ing a Garmin eTrex Summit (Fig 2)

ndash Digital topographic map in 150 000 (called ldquoN50rdquo)with glacier outlines from the AD 1980s by StatensKartverk (Norwegian mapping authorities)

wwwthe-cryospherenet32312009 The Cryosphere 3 231ndash243 2009


Fig 3 Coverage of vertical aerial photos available for the study Glacier area is from the 1980s (aerial photos Fjellanger Wideroslashe 21 July

1966 1834 B21 22 August 1976 5245 J6ndash8 29 August 1981 7084 17ndash8 44ndash46 18ndash1 4ndash5 9 12ndash14 18ndash2 6ndash7 9ndash13 18ndash3 7 9ndash11 13ndash14glacier outlines 1980s Statens Kartverk N50)

ndash Digital terrain model with 25 m resolution (DTM25) byStatens Kartverk derived from the N50 maps with a rootmean square error (RMSE) of 3ndash5 m

ndash digital glacier outlines from AD 2003 with defined iden-tification numbers (IDs) of the glaciers and basins (An-dreassen et al 2008)

ndash borders of hydrologic basins (called ldquoReginerdquo wa-tershed) the Regine watersheds have been manuallymapped by NVE with N50 as basis

The digital glacier outlines from 2003 and the 1980s weretransferred to glacier areas in GIS Digitization and calcula-tion were done using GIS-software (ArcGIS 92 byESRI)ERDAS IMAGINE 91 (by Leica Geosystems) was usedfor orthorectification and georeferencing All material wasconverted into the same projection using Universal Trans-verse Mercator (UTM) projection in World Geodetic System(WGS) 1984 European datum

42 Mapping LIA area

The glacier outlines for LIA maximum were digitized man-ually on screen using three different sources 1) the satel-lite image 2) the aerial photos and 3) the geomorphologicalmaps Glacier outlines from the 1980s and 2003 served asbasis for the minimum LIA glacier extent In the follow-ing we describe the mapping process for each of the threesources

421 Landsat image

On a glacier foreland relatively recently deglacierized vege-tation cover is absent or remains sparse Therefore the spec-tral signal is different between the glacier foreland and thearea outside (Gao and Liu 2001 Csatho et al 2005 Richardand Xiuping 2006 Albertz 2007) This spatial pattern canbe used for identifying the former LIA glacier covered areason a satellite image We displayed the Landsat image as aband 543 composite as red green and blue respectively anddigitized the LIA terminal moraine The former LIA glacierarea was in most cases adequate for manual mapping (Bau-mann et al 2008) We also tested (semi-)automatic classi-fication of the glacier foreland but overlapping spectral sig-nals between forelands and ubiquitous bare rock surfaces inthe area made it very difficult to use such classification In-clusion of other properties in the analysis eg elevation orslope enhanced the result but not to a convincing degreeTherefore we decided to digitize the LIA maximum extentmanually

Typical challenges connected to glaciers in cast shadowor with debris cover (Sidjak and Wheate 1999 Paul 2001Kaab et al 2002 Bolch and Kamp 2006 Raup et al 2007)did not cause problems in our study Shadowed areas weremostly found in the high-altitude accumulation areas and noton the low-lying glacier forelands Additionally dealing withcast shadow is easier in manual mapping than in automaticclassifications There were only few glaciers partially debris-covered and mapping of LIA moraines on the glacier fore-land was not affected by this



Fig 4 Vestre (LIA ID=118 black number) and Austre Memurubreen (LIA ID=119) with corresponding flowlines The separation of AustreMemurubreen since LIA maximum in five single glaciers until 2003 (2003 IDs=119 120 311 312 322 red numbers) is visible The flowlineon Vestre Memurubreen is divided in two branches The mean of these branches (highest to lowest elevation) was used The complete LIAglacier flowline is not visible because of coverage by the 2003 glacier flowline (directly over each other) Elevation contour interval is 100 mThe location of Vestre and Austre Memurubreen is seen in Fig 1 by letter code MEM (glacier outlines 2003 Andreassen et al 2008 digitaltopographic map Statens Kartverk N50)

422 Aerial photos

The aerial photos most of them available in stereo pairs cov-ered about 50 of the study area and 86 of the glacierizedarea (based on the glacier extent in the 1980s Fig 3) Theywere georeferenced using the digital topographical map andorthorectified in ERDAS Imagine using the DTM25 as al-titude reference following the procedure for orthorectifica-tion as described in the ERDAS manual (ERDAS IMAG-INE 2006) For each photo four to eight ground controlpoints (GCPs) depending on the fitting of the resulting im-age were collected Moraine ridges and the outline of theforelands were used to detect and map the LIA maximumextent On most glacier forelands the outermost morainescould be mapped fairly well The outermost moraine ridgewas set as LIA maximum due to the almost indisputable ab-sence of older Holocene moraines predating the LIA on theglacier forelands in this region (eg Matthews 1991 2005Shakesby et al 2008)

423 Geomorphological maps

On the geomorphological maps used in this study detailedtopographical information apart from the moraine ridges wassparse In consequence aerial photos were used to georef-erence and orthorectify them Although minor problems inorthorectification the results were considered good enoughand LIA maximum extent was manually digitized

424 Selection of LIA maximum outline

To derive one final LIA outline per glacier the different out-lines were compared The outlines of the geomorphologicalmap were assumed to be the most correct and were chosenas the final outline where available (ie at 25 glaciers) Forall other glaciers the visibility of the LIA area in the satelliteimage and aerial photo were compared and the one with theclearest visible boundary was chosen In cases with substan-tial deviation between the two outlines a closer re-check ofthe sources was made to detect any mapping errors Only thefinal outlines were used to calculate the LIA glacier area

GPS points for the outermost moraine ridges were col-lected in the field in summer 2008 at eight glaciers Theywere collected from (i) five glaciers where a geomorpholog-ical map was available and (ii) three previously unmappedglaciers The accuracy of the projection of the map in a GISsystem as well as the accuracy of the mapping itself couldbe tested using this data The difference between the GPSdata and the mapping ranged between 0 and 250 m Themean difference was 41 m with a standard deviation of 83 mThe difference at the unmapped glaciers was 83 m comparedwith 16 m at the mapped glaciers This difference mightreflect that the mapped glaciers have better visible outer-most moraine ridges Nonetheless georeferencing and map-ping were considered to be satisfactory results for all GPS-mapped glaciers with one exception

Calculation of glacier area was done using GIS SinceLIA maximum a large number of the investigated glaciers



Fig 5 Mapping of the LIA maximum extent of Styggedalsbreen (STD) on(a) the satellite image(b) an aerial photo and(c) a geomorpho-logical map Overlapping of all three sources for the sensitivity analysis with poor correspondence in(d) at Grjotbreen (GRJ) and with goodcorrespondence in(e) at Bukkeholsbreen (BUK) Location of the glaciers in Fig 1 given by letter codes (Satellite image Norsk Satellitt-dataarkivet aerial photo Fjellanger Wideroslashe 29 August 1981 7084 17ndash8 46 map Winkler 2001 glacier outline 1980s Statens KartverkN50)

separated due to glacier retreat and formed individualglaciers To reconstruct the glaciers at LIA maximum thepresent basins were adjusted to fit the formerly larger glacierareas The LIA basins had to include the transformation ofseparated glaciers into one basin if they constituted a sin-gle glacier during LIA maximum The boundary of singleglacier units at LIA maximum was derived by clipping theLIA maximum extent with the LIA basins and glacier areaswere calculated automatically using GIS Finally all glacierssmaller than 001 km2 were deleted from the inventory of theglaciers at LIA maximum At this scale it is difficult to distin-guish between glaciers and perennial snowfields (eg Paul2009) Furthermore the resolution and accuracy of mappingis restricted by the pixel size of the satellite image This sizeclass was also not included in the inventory of 2003 (An-dreassen et al 2008)

43 Creating LIA centrelines

To calculate glacier lengths during LIA maximum centre-lines were digitized manually using the LIA glacier outlinesand the contour lines of N50 as basis Flowlines were dig-itized perpendicular to the contour lines preferably in themiddle of the glacier and from its highest to the lowest points(Paul 2009) On glaciers with different tributaries or wide

accumulation areas more than one flowline was digitized(eg Memurubreen in Fig 4) In such cases a mean forthe flowlines was calculated and hereafter used as ldquoglacierlengthrdquo The number of flowlines varied between one andthree at individual glaciers

44 Creating LIA inventory data

Inventory data of the glaciers at LIA maximum (LIA in-ventory) were calculated automatically using ArcGIS andthe DTM25 The calculated inventory data included mini-mum maximum and mean altitude slope and aspect Ex-cept for aspect all variables were derived directly from Ar-cGIS The aspect was calculated using a routine described inPaul (2007)

45 Sensitivity analysis

The glacier outlines were as mentioned derived from threedifferent sources satellite images aerial photographs andgeomorphological maps For 18 glaciers all three sourceswere available and compared (examples in Fig 5) The de-rived areas of all three sources compared well the coefficientof variation was 17 and all areas were within the 95-confidencel interval The differences between the glacierareas from the satellite image and the aerial photos were


S Baumann et al Mapping glaciers in Jotunheimen South-Norway 237

Table 1 Classification of the glacier area during LIA maximum into size intervals and comparison with 2003

Area Interval Number [n] Number [] Mean area [km] Cumulative [][km2] LIA 2003 LIA 2003 LIA 2003 LIA 2003

lt01 26 75 1116 2863 006 006 056 229[01 05) 96 101 4120 3855 026 024 909 1432[05 10) 37 36 1588 1374 073 072 1842 2771[10 50) 63 43 2704 1641 235 220 6947 7695[50 100) 8 7 343 267 687 640 8842 10000ge100 3 ndash 129 ndash 1119 ndash 10000 10000Total 233 262 100 100 124 087 ndash ndash

small but mapping using the satellite image revealed a closeragreement to the geomorphological maps than mapping us-ing aerial photos Both the Landsat image and aerial photosgave larger areas than the geomorphological maps 13 km2

and 22 km2 respectively for the total of 18 glaciers an in-crease of 001 or 002 respectively In conclusion thecomparison of the mapping sources showed good agreementand the derived LIA glacier outlines were considered to berobust for all three methods

5 Results

We mapped the LIA maximum extent of 233 glaciers in Jo-tunheimen with a total area of about 290 km2 (Fig 1) Theindividual parts of composite glaciers and ice caps are herereferred to as single glaciers The mean glacier area was124 km2 The size distribution of the glaciers is shown inTable 1 Most of the glaciers fall within the interval [0105) km2 whereas the glaciers in the interval [10 50) km2

exhibit the largest contribution to the total area Despite alarge percentage of the overall number of glaciers (68) thegroup of smallest glaciers represents only a minor part ofthe total glacier area (18) By contrast less than 5 of allglaciers (the largest class) constitute more than 30 of the to-tal glacier area Only three glaciers were larger than 10 km2

at LIA maximum The largest glacier in Jotunheimen wasOslashstre Memurubreen with an area of 124 km2

The maximum altitude of the glaciers ranged between1500 and 2500 m asl with a mean of 2010 m asl (stan-dard deviation (σ)=170 m asl) the minimum altitude be-tween 1000 and 2400 m asl with a mean of 1590 m asl(σ=206 m asl) The highest elevation range was found onStyggebreen with a range of 1396 m The minimum maxi-mum mean and median altitude all generally increased fromWest to East A relation to the glacier size was only de-tectable for the minimum altitude The plot of minimumglacier altitude vs area (Fig 6) revealed a lower minimumaltitude for larger glaciers than for smaller ones The coeffi-cient of determination of the best fitting straight line throughall points is 036 The coefficient of determination for themean maximum glacier altitudes shown in the same plot

Fig 6

Fig 6 Scatter plot of minimum (blue dots) and maximum (greendots) elevation vs LIA glacier area The solid lines give the best fit-ting straight line of minimum and maximum elevation respectivelyNote logarithmic horizontal scale

(Fig 6) is 026 Scatter plots of aspect vs glacier size and ofslope vs aspect showed no distinct pattern Furthermore norelationship was found between the four calculated altitudes(minimum maximum mean and median) and mean aspector slope

The length of the glaciers varied between 134 and 6818 mwith a mean of 1554 m About two thirds (656) of theglaciers had lengths shorter than the mean The median valuewas 1064 m More than half of the lengths (502) were inthe interval [10 50) km (Table 2) Only eight glaciers hada length exceeding 50 km Soslashre Veobreen had the longestflowline with a length of 6818 m 41 glaciers (18) had morethan one flowline

The LIA inventory data were compared with the glacierdata from AD 2003 The total area declined from about290 km2 during LIA maximum to 190 km2 in AD 2003(minus35) The relative area reduction between LIA maximumand 2003 is slightly higher for smaller glaciers than for larger


238 S Baumann et al Mapping glaciers in Jotunheimen South-Norway

Table 2 Classification of the glacier length during LIA maximuminto length intervals and comparison with 2003

Length Number [n] Number [] Mean length [km]interval [km] LIA 2003 LIA 2003 LIA 2003

lt05 38 92 1631 3511 033 033[05 10) 70 85 3004 3244 073 071[10 50) 117 85 5021 3244 213 208ge50 8 ndash 343 ndash 616 ndashTotal 233 262 100 100 155 102

ones but without indicating any clear pattern Howeverthe range of change for the smaller glaciers is large (0ndash(minus100)) The highest reduction (minus47) is found in theinterval [01 05) km2 the smallest (minus28) in the interval[50 100) km2 (Fig 7) No notable spatial pattern in area re-duction could be detected between LIA maximum and 2003(Fig 8) The spatial pattern of higher relative area reductionin the northeastern and eastern part of Jotunheimen betweenthe 1980s and 2003 detected by Andreassen et al (2008) isnot visible between LIA and 2003 The calculated hypsogra-phy for 100 m intervals of LIA maximum and 2003 (Fig 9)shows not surprisingly a higher absolute area loss at lowerelevations The highest absolute loss occurred at an elevationrange of 1500ndash1700 m asl The increase in mean elevation(+61 m) is slightly higher than for median elevation (+55 m)

As shown on Fig 10 a higher number of glaciers smallerthan 01 km2 existed in AD 2003 but none larger than100 km2 Except for these two intervals the relative dis-tribution of glaciers was similar at both times In to-tal 13 glaciers disappeared between LIA and 2003 and34 glaciers separated into two or more parts

The four altitude values (minimum maximum mean andmedian) of LIA maximum were compared with the respec-tive values of 2003 The resulting differences showed neitherany spatial pattern nor any relation to glacier size mean as-pect or slope

The mean length was reduced by 34 from LIA maxi-mum (16 km) to 2003 (10 km) The number of glacierswith lengths in the intervallt05 km decreased from theLIA maximum to 2003 (Table 2) The relative number ofglacier length stayed nearly the same in the size interval [0510) km and declined from 50 to 32 in the interval [1005) km

6 Discussion

61 Comparison of sources

For about one third of the aerial photos the accuracy of fit-ting was not satisfying Close examination of the sourcesrevealed differences between contour lines from the topo-graphical map (N50) and the DTM25 (mean altitudinal er-

Fig 7

Fig 7 Scatter plot of relative area change of glacier area between2003 and LIA vs LIA glacier area The solid lines give mean valuesper size class the bold dashed line shows the mean for all glaciers(raw data glacier outlines 2003 Andreassen et al 2008)

ror=33 mσ=126 m) Those were possibly related to inter-polation errors To quantify the impact of this differenceseveral aerial photos were orthorectified again using onlythe DTM25 The LIA areas digitized with these orthopho-tos were compared with the first results The coefficient ofdetermination was close to 1 (r2=09997) Thus the error re-garding the resulting LIA areas could be ignored and errorsin orthorectification of the aerial photos were assumed to benegligible The reported difference in fitting might be causedby the altitude range on each aerial photo (often exceeding1000 m) The determined deviation is comparable with otherstudies (eg Csatho et al 2005)

In several cases the geomorphological maps were obvi-ously drawn using non-orthorectified aerial photos For geo-referencing of these maps aerial photos had to be used Onlyon those the precise position of the moraine ridges couldbe mapped if no other useful information was available onthe maps Therefore the geomorphological maps cannot beregarded as an independent source To determine the po-tential error emerging during this procedure the GPS-datacollected in field were used The difference between theseGPS-coordinates and the position of the outermost moraineon geomorphological maps was calculated for five glaciersThe mean error of the resulting LIA area is about 282 m2

(σ=88 m2) This error is smaller than the pixel size ofthe satellite image (30 mtimes30 m) The RMSE of the geo-referenced satellite image itself is about 065 pixels whichis about 20 m (Andreassen et al 2008) The error result-ing from topographical inaccuracy of the geomorphologicalmaps is therefore acceptable for our purpose



Fig 8 Colour-coded relative area change between LIA maximum and 2003 for each single glacier Shown are glacier extents of LIAmaximum (raw data glacier outlines 2003 Andreassen et al 2008)

Fig 9

Fig 9 Glacier hypsography with 100 m contour interval for LIAmaximum (grey) and 2003 (yellow) including statistical data LIAarea-altitude distribution is an assumption based on the topographicmap of the 1980s (raw data glacier outlines 2003 Andreassen et al2008)

62 Uncertainties in mapping the LIA maximum

When mapping the LIA maximum outlines the outermostmoraine was assumed to represent the maximum glacier ex-tent during LIA This assumption was based on extensiveprevious studies (see Sect 3) but exceptions might be ice-cored moraine systems at few high-altitude cirque glaciers inEast Jotunheimen eg at Grasubreen (Oslashstrem 1964) Theiroutermost ridges could pre-date the LIA although evidenceis yet sparse (eg Oslashstrem 1964 Winkler 2001 Shakesby etal 2004 2008) Because of short distances between indi-vidual ridges the possible error by non-exact location of theridge representing the outermost LIA position within these

Fig 10

Fig 10 Comparison of area intervals between LIA maximum and2003 Both number and area sum are compared (raw data 2003Andreassen et al 2008)

complex moraine systems are considered to be small AtGrasubreen a comparison of the LIA glacier area using sixdifferent moraine walls as LIA glacier limit showed a coef-ficient of variance of 45

Another uncertainty is the unknown LIA maximum to-pography of the glacier which affects the accuracy of theLIA inventory parameters The slope calculation using thepresent deglaciated glacier foreland differs from the slopeof the former glacier tongue for example Estimations ofmean median and maximum altitude will also be influencedby the lack of a LIA glacier surface topography The drainagebasins of larger basins might also be different at the LIAmaximum but this is difficult to account for in the analysesReconstructions of LIA surfaces could be possible but willnevertheless be an estimate as no reliable maps are availablefor validation

As mentioned in Sect 42 all glaciers smaller than001 km2 were removed from the inventory because of theuncertainties in mapping In particular it is difficult to



distinguish perennial snow patches from small glaciers onthat scale (Paul and Andreassen 2009) Using a largerglacier minimum size revealed that an exclusion of the small-est area interval would indeed alter the results In theoriginal dataset 13 glaciers disappeared between LIA max-imum and 2003 Nine glaciers disappeared between LIA and2003 if all glacierslt01 km2 at LIA maximum are excludedThese nine glaciers ranged between 011 and 025 km2 atLIA maximum In relative numbers concerning the totalglacier number 56 of the glaciers disappeared with theoriginal dataset and 50 with LIA and 2003 areagt01 km2

Mapping should preferably be an objective process withclear rules and a high degree of reproducibility independentof the analyst In reality however different analysts willas a result of subjective interpretation never produce the ex-actly same results (eg Andreassen et al 2008) Thereforeincluding any kind of objective analysis would be desirableThe test of using supervised or unsupervised image classifi-cation was unfortunately not successful in this study

For the inventory glacier length was derived by using themean of all tributaries to compare lengths at the LIA max-imum and 2003 The results represent those glaciers asa whole but the purely theoretical character of this valueshould be kept in mind as using the mean glacier length sup-presses the individual signal of the glaciers By selectingthe maximum flowline a length that actually is measured onthe glacier could have been chosen But regarding branchedflowlines the change in maximum value for two points oftime might cause a change of the corresponding tributaryThe flowlines would not longer be identical and a compari-son would be impossible Changes in the mean value refer toall tributaries and remain therefore comparable Thereforecalculating the mean glacier length from the 1ndash3 flowlinesfor each glacier was considered to be an appropriate methodfor recording LIA glacier lengths

The mapping of LIA glacier outlines using aerial photoshad some limitations As already mentioned in Sect 42only 86 of the glacierized area was covered by aerial pho-tos (Fig 3) Furthermore due to distortion and displace-ment towards the edges of the individual photos the effectiveworking size was even smaller than the whole photo Finallyit was sometimes difficult to detect the moraine ridges incases where stereo pairs were missing A general disadvan-tage of using aerial photos is the time required for prepara-tion Each single photo had to be referenced separately De-pending on the number of photos needed to cover the studyarea the process of orthorectification and georeferencing ofthe aerial photos took much longer in total compared withprocessing the satellite image Despite multiple use of someGCPs the processing time was almost the same for eachphoto The big advantage of the aerial photos however istheir high resolution (04 mtimes04 m) Objects could be iden-tified very precisely whereas the resolution of the satelliteimage (30 mtimes30 m) is a limiting factor

63 Comparison with other regions

In the following we compare the LIA results obtained for Jo-tunheimen with similar inventories from the European Alpsthe Southern Alps of New Zealand and the Canadian Arc-tic Unfortunately no LIA inventory on a regional scale forother glacier regions of Norway is available yet The Eu-ropean Alps and the Southern Alps of New Zealand have ahigh-alpine character with mostly individual glaciers simi-lar to Jotunheimen (eg Chinn 2001 Zemp et al 2008)The glaciers on Cumberland Peninsula of Baffin Island inthe Canadian Arctic are also individual glaciers but mainlyof larger sizes (Paul and Kaab 2005)

The glacier inventory of LIA maximum in Switzerlandwas compiled by Maisch et al (2000) The area decreasefrom AD 1850 (LIA maximum) until 1999 is 34 perdecade (Paul et al 2004b) or 51 in total This decreaseis higher than the corresponding one between LIA maximum(AD 1750) and 2003 in Jotunheimen (sim35 sim13 perdecade) Glaciers in the two regions Jotunheimen and theSwiss Alps show an asynchronous pattern of the influenceof North Atlantic Oscillation (NAO) modes (Nesje and Dahl2003) and poorly correlating annual net balances (Guntherand Widlewski 1986 Steiner et al 2008) Whereas massbalance in southern Norway is well correlated to winterprecipitation (Nesje and Dahl 2003 Matthews and Briffa2005) mass balance in the Alps does no show such an im-pact (Bohm et al 2007) As a consequence the increase intemperature affected the European Alps more strongly thanJotunheimen A higher relative loss at small glaciers also oc-curred in Switzerland (Paul et al 2007) Such a differencein relative area decrease depending on glacier size was notdetected in Jotunheimen Whereas 69 of the glacier areain Jotunheimen is due to glacierslt5 km2 the correspondingfigure for the Swiss Alps is only 55 (Zemp et al 2008)At LIA maximum a considerably higher amount of glacierarea was due to glaciersgt10 km2 in Switzerland (33) thanin Jotunheimen (12)

A glacier inventory for LIA maximum (AD 1850) existsfor the Austrian Alps (Gross 1987) A second inventory wascompiled for AD 1979 (Rott et al 1993) The glacier areadecrease between 1850 and 1969 was 46 (Gross 1987)This area loss is much higher than calculated for the SwissAlps (27) in the comparable period 1850ndash1973 (Zemp etal 2008) and for Jotunheimen for the period 1750ndash2003(35) The relative area loss is higher for small glaciers andSouth and West exposition (Gross 1987) Glaciers in theAustrian Alps exhibit relatively smaller sizes compared withJotunheimen though the maximum extent is higher Theglacier size is possibly the reason for the higher decrease inglacier area as small glaciers will respond more sensitive tochanges in climate (Haeberli 1995 Kuhn 1995 Bohm etal 2007)



The Southern Alps of New Zealand are a mountain sys-tem with a higher number of individual glaciers (gt3000)and a larger total glacier area compared with Jotunheimen(1160 km2 in AD 1978) New Zealand exhibits a pronouncedmaritime climate The mass balance is strongly influencedby atmospheric circulation patterns and correlates well withboth the Southern Oscillation Index (SOI) and the PacificDecadal Oscillation (PDO) mode (Chinn et al 2005) Theglacier inventory data for LIA maximum (owing to the lackof reliable widespread datings set to AD 1850 when manyglaciers remained close to their LIA maximum position)were compiled by detection of moraines on vertical aerialphotos (Chinn 1996) A second inventory of the whole re-gion exists for AD 1978 Many of the larger valley glaciersare debris-covered and recently form glacier fronts calvinginto pro-glacial lakes (Chinn 1996) During LIA maximum67 of the glaciers were in the area class interval [1 5)km2 (in total 98lt5 km2) In 1978 90 of the glaciersare smaller than 05 km2 a shift towards smaller area inter-vals compared with LIA maximum (M Hoelzle personalcommunication 2009) The area decrease from LIA max-imum until 1978 is 49 (39 per decade) (Hoelzle et al2007) In the 1980s and 1990s many debris-free glaciersadvanced (Chinn et al 2005) but retreat continued at mostof the debris-covered glaciers especially where proglaciallakes developed The debris-covered glaciers with proglaciallakes have generally experienced massive mass loss duringrecent years (Chinn et al 2008) The relative area distri-bution in the Southern Alps and Jotunheimen is similar forLIA maximum The reduction in glacier area since regionalLIA maximum is smaller in Jotunheimen (until 2003) thanin the Southern Alps of New Zealand (until 1978) Between1981 and 2003 many of the larger glaciers in Jotunheimenexperienced only small reductions in area However glaciersin the eastern and southeastern Jotunheimen had a notablereduction (Andreassen et al 2008) Maritime glaciers areregarded as highly climate sensitive (Kuhn 1984 Laumannand Reeh 1993 Dyurgerov and Meier 1999) This mightexplain the differences between the two regions

LIA maximum glacier extent on Baffin Island in the Cana-dian Arctic (occurring about AD 1920s Paul and Kaab2005) was mapped by trimline and moraine survey usingremote sensing (Paul and Kaab 2005 Paul and Svoboda2009) The glaciers at LIA maximum were larger comparedwith Jotunheimen and only 19 werelt5 km2 (Paul andSvoboda 2009) On Cumberland Peninsula a part of Baf-fin Island there is no scatter towards smaller glaciers and adependency of relative area change on glacier size (Paul andKaab 2005) Since LIA maximum the glaciers on BaffinIsland have lost 13 (16 per decade Paul and Svoboda2009) on Cumberland Peninsula 11 (14 per decade Pauland Kaab 2005) The difference in glacier sizes and climatesetting to Jotunheimen limit however the value of this com-parison

7 Conclusions

This study showed that satellite imagery and aerial photoscould be used for manual mapping of glacier outlines at LIAmaximum in Jotunheimen on a regional scale and deliveringinventory data of LIA maximum Glacier flowlines were dig-itized manually for all glaciers to derive LIA glacier lengthsChallenges occurred during the mapping process (eg or-thorectification of the aerial photos combination of the dif-ferent topographical sources) but the reliability of the map-ping results are considered to be satisfactory

Only a few glaciers vanished completely between LIA and2003 Overall the glaciers were larger at LIA maximum es-pecially at the upper and lower end of the glacier size rangeThe relative glacier area reduction in Jotunheimen (35) isnot as strong as in other regions (Swiss Alps 51 SouthernAlpsNew Zealand 49) It has to be noted that the timingof the LIA maximum differs between the regions

A comparison of the LIA inventory with the glacier inven-tories of the 1960s and the 1980s of Jotunheimen could beinteresting to study glacier behaviour in relation to changingclimate parameters The methods used here could be appliedto map the LIA maximum glacier extent in other regions ofNorway

AcknowledgementsWe thank the three reviewers Andreas KaabIan Brown and Martin Hoelzle for their useful suggestions ona former version of the manuscript and also Mauri Pelto andFrank Paul for their helpful comments Remarks and language re-vision from L A Rasmussen (University of Washington) improvedthe manuscript This study was supported by a personal grantfrom theBayerischen Graduiertenforderung nach dem BayerischenEliteforderungsgesetz(BayEFG) by a grant for fieldwork in 2008 inNorway from theUniversitatsbund der Universitat Wurzburgand apersonal grant from theQualifikationsprogramm fur Wissenschaft-lerinnen an der Universitat Wurzburg The study is connectedto the DFG-funded project MaMoGla (Grant WI-17013) and NVE

Edited by J O Hagen

References

Albertz J Einfuhrung in die Fernerkundung 3 ed Wis-senschaftliche Buchgesellschaft Darmstadt 2007

Andreassen L M Elvehoslashy H Kjoslashllmoen B Engeset R Vand Haakensen N Glacier mass-balance and length variation inNorway Ann Glaciol 42 317ndash325 2005

Andreassen L M Paul F Kaab A and Hausberg J E Landsat-derived glacier inventory for Jotunheimen Norway and deducedglacier changes since the 1930s The Cryosphere 2 131ndash1452008httpwwwthe-cryosphere-discussnet21312008

Baumann S Andreassen L M and Winkler S Mapping Lit-tle Ice Age glacier maxima in Jotunheimen using aerial pho-tography and Landsat imagery Proceedings of EARSeL-LISSIGWorkshop on Remote Sensing of Snow and Glaciers in Bern on-line available athttpwwwearselorgworkshopsLISSIG2008BaumannPaperpdf 2008



Bickerton R W and Matthews J A ldquoLittle Ice Agerdquo varia-tions of outlet glaciers from the Jostedalsbreen ice-cap southernNorway a regional lichenometric-dating study of ice-marginalmoraine sequences and their climatic significance J QuaternarySci 8 45ndash66 1993

Bohm R Schoner W Auer I Hynek B Kroisleitner C andWeyss G Gletscher im Klimawandel Zentralanstalt fur Mete-orologie und Geodynamik Wien 2007

Bogen J Wold B and Oslashstrem G Historic glacier variationsin Scandinavia in Glacier Fluctuations and Climatic Changeedited by Oerlemans J Kluwer Academic Publishers Dor-drecht 109ndash128 1989

Bolch T and Kamp U Glacier Mapping in High Mountains UsingDEMs Landsat and Aster Data Grazer Schriften der Geographieund Raumforschung 41 37ndash48 2006

Bolch T Buchroithner M Pieczonka T and Kunert A Plani-metric and volumetric glacier changes in the Khumbu HimalNepal since 1962 using Corona Landsat TM and ASTER dataJ Glaciol 54 592ndash600 2008

Chinn T New Zealand glacier responses to climate change of thepast century New Zeal J Geol Geop 39 415ndash428 1996

Chinn T J Distribution of the glacial water resources of NewZealand J Hydrol (NZ) 40 139ndash187 2001

Chinn T J H Salinger J Fitzharris B B and Willsman AGlaciers and climate Bulletin of the Federal Mountain Clubs ofNZ 171 1ndash15 2008

Chinn T J H Winkler S Salinger M J and Haakensen NRecent glacier advances in Norway and New Zealand ndash a com-parison for their glaciological and meteorological causes GeogrAnn A 87 141ndash157 2005

Csatho B M van der Veen C J and Tremper C MTrimline mapping from multispectral Landsat ETM+ imageryGeographie Physique et Quaternaire 59 49ndash62 2005

De Jong R Hammarlund D and Nesje A Late Holocene effec-tive precipitation variations in the maritime regions of south-westScandinavia Quaternary Sci Rev 28 54ndash64 2009

Dyurgerov M B and Meier M F Analysis of winter and summerglacier mass balance Geogr Ann A 81 541ndash554 1999

ERDAS IMAGINE ERDAS IMAGINE Tour Guides 730 pp2006

Erikstad L and Sollid J L Neoglaciation in South Norway us-ing lichenometric methods Norsk Geogr Tidsskr 40 85ndash1051986

Faeliggri K Brevariasjoner i Vestnorge i de siste 200ar Nature 72230ndash243 1948

Gao J and Liu Y Applications of remote sensing GIS and GPSin glaciology a review Prog Phys Geog 25 520ndash540 2001

Gross G Der Flachenverlust der Gletscher inOsterreich 1850 ndash1920ndash1969 Zeitschrift fur Gletscherkunde und Glazialgeologie23 131ndash141 1987

Gunther R and Widlewski D Die Korrelation verschiedener Kli-maelemente mit dem Massenhaushalt alpiner und skandinavis-cher Gletscher Zeitschrift fur Gletscherkunde und Glazialgeolo-gie 22 125ndash147 1986

Haeberli W Glacier fluctuations and climate change detection- operational elements a worldwide monitoring strategy WMOBulletin 44 23ndash31 1995

Hoel A and Werenskiold W Glaciers and snowfields in NorwayNorsk Polarinstitutt Skrifter Nr 114 Oslo University Press

Oslo 1962Hoelzle M Chinn T Stumm D Paul F Zemp M and Hae-

berli W The application of inventory data for estimating pastclimate change effects on mountain glaciers A comparison be-tween the European Alps and the Southern Alps of New ZealandGlobal Planet Change 56 69ndash82 2007

Innes J L Lichenometry Prog Phys Geog 9 187ndash254 1985IPCC Climate Change 2007 The Physical Science Basis Con-

tribution of Working Group I to the Fourth Assessment Reportof the Intergovernmental Panel on Climate Change edited bySolomon S Qin D Manning M Chen Z Marquis M Av-eryt K B Tignor M and Miller H L Cambridge UniversityPress Cambridge New York 2007

Kaab A Huggel C Paul F Wessel R Raup B Kieffer H andKargel J S Glacier Monitoring from ASTER Imagery Ac-curacy and Applications in Proceedings of EARSeL-LISSIG-Workshop Observing our Cryosphere from Space Bern 43ndash532002

Kargel J S Abrams M J Bishop M P Bush A Hamiton GJiskoot H Kaab A Kieffer H H Lee E M Paul F RauF Raup B Shroder J F Soltesz D Stainforth D StearnsL and Wessel R Multispectral imaging contributions to globalland ice measurements from space Remote Sens Environ 99187ndash219 2005

Kuhn M Mass balance imbalances as criterion for a climatic clas-sification of glaciers GeogrAnn A 66 229ndash238 1984

Kuhn M The mass balance of very small glaciers Zeitschrift furGletscherkunde und Glazialgeologie 31 171ndash179 1995

Laumann T and Reeh N Sensitivity to climate change of themass balance of glaciers in southern Norway J Glaciol 39656ndash665 1993

Maisch M Wipf A Denneler B Battaglia J and Benz CDie Gletscher der Schweizer Alpen Gletscherhochstand 1850Aktuelle Vergletscherung Gletscherschwund Szenarien 2 ededited by NFP 31 vdf Hochschulverlag an der ETH Zurich2000

Matthews J A Families of lichenometric dating curves from theStorbreen gletschervorfeld Jotunheimen Norway Norsk GeogrTidsskr 28 215ndash235 1974


Matthews J A Experiments on the reproducibility and reliabilityof lichenometric dates Storbreen gletschervorfeld JotunheimenNorway Norsk Geogr Tidsskr 29 97ndash109 1975

Matthews J A A lichenometric test of the 1750 end-moraine hy-pothesis Storbreen gletschervorfeld southern Norway NorskGeogr Tidsskr 31 129ndash136 1977

Matthews J A The late Neoglacial (ldquoLittle Ice Agerdquo) glacier max-imum in southern Norway new14C-dating evidence and cli-matic implications Holocene 1 219ndash233 1991

Matthews J A ldquoLittle Ice Agerdquo glacier variations in Jotunheimensouthern Norway a study in regionally controlled lichenometricdating of recessional moraines with implications for climate andlichen growth rates Holocene 15 1ndash19 2005

Matthews J A and Briffa K R The rsquoLittle Ice Agersquo re-evaluation of an evolving concept Geogr Ann A 1 17ndash362005

Matthews J A and Dresser P Q Holocene glacier variation



chronology of the Smoslashrstabbtindan massif Jotunheimen south-ern Norway and the recognition of century- to millennial-scaleEuropean Neoglacial Events Holocene 18 181ndash201 2008

Nesje A Latest Pleistocene and Holocene alpine glacier fluc-tuations in Scandinavia Quaternary Sci Rev 28 2119ndash2136doi101016jquascirev200812016 2009

Nesje A and Dahl S O The lsquoLittle Ice Agelsquo- only temperatureHolocene 13 139ndash145 doi1011910959683603hl603fa 2003

Nesje A Bakke J Dahl S O Lie Oslash and Matthews J A Nor-wegian mountain glaciers in the past present and future GlobalPlanet Change 60 10ndash27 2008a

Nesje A Dahl S O Thun T and Nordli Oslash The ldquoLittle IceAgerdquo glacial expansion in western Scandinavia summer tem-perature or winter precipitation Clim Dynam 30 789ndash801doi101007s00382-007-0324-z 2008b

Nordli Oslash Lie Oslash Nesje A and Benestad R Glacier mass bal-ance in southern Norway modelled by circulation indices andspring-summer temperature AD 1781ndash2000 Geogr Ann A 87431ndash445 2005

Oslashstrem G Ice-cored moraines in Scandinavia Geogr AnnXLVI 3 282ndash337 1964

Oslashstrem G and Ziegler T Atlas over breer i Soslashr-Norge NVEOslo 1969

Oslashstrem G Dale Selvig K and Tandberg K Atlas over breer iSoslashr-Norge NVE Oslo 1988

Oslashyen P A Isbraeligstudier i Jotunheimen Nytt Magasin for Naturv-idenskabere 31 26ndash27 1893

Paul F Evaluation of different methods for glacier mapping usingLandsat TM in Proceedings of EARSeL-SIG Workshop LandIce and Snow Dresden 239ndash245 2001

Paul F The New Swiss Glacier Inventory 2000 Application ofRemote Sensing and GIS Schriftenreihe Physische GeographieGlaziologie und Geomorphodynamik Geographisches Institutder Universitat Zurich Zurich 210 pp 2007

Paul F Guidelines fort he compilation of glacier inventorydata form digital sources GLIMS online available athttpglobglacierchdocsguidelinesinventorypdf 2009

Paul F and Andreassen L M Creating a glacier inventory forthe Svartisen region (Norway) from Landsat ETM+ satellite dataChallenges and results J Glaciol 55 607ndash618 2009

Paul F and Kaab A Perspectives on the production of a glacier in-ventory from multispectral satellite data in Arctic Canada Cum-berland Peninsula Baffin Island Ann Glaciol 42 59ndash66 2005

Paul F and Svoboda F A new glacier inventory on southern Baf-fin Island Canada from ASTER data II Data analysis glacierchange and applications Ann Glaciol 50 22ndash31 2009

Paul F Huggel C and Kaab A Combining satellite multispec-tral image data and a digital elevation model for mapping debris-covered glaciers Remote Sens Environ 89 510ndash518 2004a

Paul F Huggel C Kaab A Kellenberger T and Maisch MComparison of TM-derived glacier areas with higher resolutiondata sets in Proceedings of EARSeL-LISSIG-Workshop Ob-serving our Cryosphere from Space Bern 15ndash21 2003

Paul F Kaab A and Haeberli W Recent glacier changes in theAlps observed by satellite Consequences for future monitoringstrategies Global Planet Change 56 111ndash122 2007

Paul F Kaab A Maisch M Kellenberger T and Hae-berli W Rapid disintegration of Alpine glaciers ob-served with satellite data Geophys Res Lett 31 L21402

doi1010292004GL020816 2004bRaup B Kaab A Kargel J S Bishop M P Hamiton G Lee

E Paul F Rau F Soltesz D Khalsa S J S Beedle Mand Helm C Remote sensing and GIS technology in the GlobalLand Ice Measurements from Space (GLIMS) Project ComputGeosci 33 104ndash125 2007

Richard J A and Xiuping J Remote Sensing Digital Image Anal-ysis An Introduction 4 ed Springer Verlag Berlin 2006

Rott H Scherler K E Reynaud L Barbero R S and ZanonG Glaciers of Europe - Glaciers of the Alps in Satellite Im-age Atlas of glaciers of the World edited by Williams R S Jand Ferringo J G USGS Professional Paper 1386-E-1 USGS1993

Shakesby R A Matthews J A and Schnabel C Cosmogenic10Be and26Al ages of Holocene moraines in southern Norway IIevidence for individualistic responses of high-altitude glaciers tomillennial-scale climatic fluctuations Holocene 18 1165ndash11772008

Shakesby R A Matthews J A and Winkler S Glaciervariations in Breheimen southern Norway relative-age datingof Holocene moraine complexes at six high-altitude glaciersHolocene 14 899ndash910 2004

Sidjak R W and Wheate R D Glacier mapping of the Illecille-waet icefield British Columbia Canada using Landsat TM anddigital elevation data Int J Remote Sens 20 273ndash284 1999

Solomina O Barry R and Bodnya M The retreat of tienshan glaciers (Kyrgyzstan) since the Little Ice Age estimatedfrom aerial photographs lichenometric and historical data Ge-ogr Ann A 86 205ndash215 2004

Statens Kartverk Galdhoslashpiggen Norway Statens Kartverk Oslo1985

WGMS Global Glacier Changes facts and figures edited byZemp M Roer I Kaab A Hoelzle M Paul F and HaeberliW UNEP World Glacier Monitoring Service Zurich 88 pp2008

Winkler S Untersuchungen zur Klima- und Morphodynamikin skandinavischen Gebirgsregionen wahrend des Holozanndash ein Vergleich ihrer Wechselwirkungen und Prozeszligsys-teme im uberregionalen Kontext kaltgemaszligigter maritimerGebirgsregionen Habilitation thesis Fachbereich Geogra-phieGeowissenschaften Universitat Trier Trier 2001

Winkler S Von der ldquoKleinen Eiszeitrdquo zum ldquoglobalenGletscherruckzugrdquo Eignen sich Gletscher als Klimazeu-gen Colloquia Academia Abhandlungen der Mathematisch-naturwissenschaftlichen Klasse 2002 Nr 3 Akademie derWissenschaften und der Literatur MainzFranz Steiner VerlagStuttgart 2002

Winkler S Gletscher und ihre Landschaften ndash eine illustri-erte Einfuhrung Wissenschaftliche Buchgesellschaft Darmstadt2009

Wolken G J High-resolution multispectral techniques for map-ping former Little Ice Age terrestrial ice cover in the CanadianHigh Arctic Remote Sens Environ 101 104ndash114 2006

Zemp M Paul F Hoelzle M and Haeberli W Glacier fluctua-tions in the European Alps 1850-2000 an overview and spatio-temporal analysis of available data in The darkening peaksGlacial retreat in scientific and social context edited by OrloveB Wiegandt E and Luckman B University of CaliforniaPress 2008



Fig 1 Location of the study area Jotunheimen (inset) and glaciers (gt001 km2) with flowlines during the LIA maximum Letter codesdenote STD=Styggedalsbreen BUK=Bukkeholsbreen GRJ=Grjotbreen MEM=Memurubreen (Inset map ESRI Templates)

2 Study area

The study area of Jotunheimen is located in central SouthNorway (615 N 83 E) (Fig 1) It has a high-alpinecharacter and the highest points of Norway are locatedhere (Galdhoslashpiggen 2469 m asl Glittertind 2464 m asl)The present glaciers (ie AD 2003) are mostly small in-dividual valley-type and cirque-type glaciers separated bysteep rock-walls and range from about 1300 to 2300 m asl(Andreassen et al 2008) Long-term recorded mass bal-ance along a West-East profile in southern Norway revealsa strong gradient in mass balance with decreasing massturnover towards the more continental drier interior (egOslashstrem et al 1988)

The study area lies in a transitional zone between maritimeand continental climate and represents the most continentalglacier area in southern Norway (Oslashstrem et al 1988)

3 Timing of the LIA maximum in Jotunheimen

In contrast to Jostedalsbreen to the West there are no his-torical documents or images of the glacier area of Jotun-heimen that would allow a precise dating of the LIA max-imum Oslashyen (1893) and Hoel and Werenskiold (1962) re-spectively report only a vague eyewitness story about themaximum extent of Storbreen but no clear evidence thatcould be used for a distinct timing of the LIA maximumTherefore existing reconstructions of the glacial chronol-ogy in Jotunheimen prior to the first scientific measurementsand historical photo-record around AD 1900 are primarilybased on lichenometry supported by14C-datings lake sedi-ment analysis mire and colluvial profiles etc (eg Matthews

1974 1975 1977 2005 Innes 1985 Erikstad and Sollid1986 Winkler 2001 Matthews and Dresser 2008) Thesestudies suggest that the timing of the culmination of the LIAin Jotunheimen falls roughly between AD 1750 and 1800In West and Central Jotunheimen the outermost moraines ofthe LIA often date from around AD 1750 whereas in EastJotunheimen the related moraines tend to be a few decadesyounger ie date from around AD 17801800 (Fig 2 Win-kler 2002 Matthews 2005) Owing to the regional ecolog-ical conditions (glacier forelands above the tree lines rel-atively stable moraine ridges limited press on lichens byvascular plants and mosses usually sufficient humidity forlichen growth siliceous rock types etc) lichenometry has acomparable high temporal resolution and gives reliable re-sults (Matthews 2005) The accuracy of lichenometry indating the LIA maximum in this region are assumed to bewithin plusmn20 years for the more recent sophisticated studiesusing regional lichen growth curves (eg Matthews 2005)

The AD 1750 LIA maximum in West and Central Jotun-heimen is interpreted analogously to the mid-18th centuryLIA maximum at Jostedalsbreen (Bickerton and Matthews1993 Nesje et al 2008a) The primary climatologic causefor this advance is considered to be increased winter precipi-tation (Nesje and Dahl 2003 Nesje et al 2008a b De Jonget al 2009 Nesje 2009) By contrast the second half ofthe 18th century was dominated by below-average summertemperatures (Winkler 2001 Nordli et al 2005) In this pe-riod the outlets of Jostedalsbreen retreated slowly from theiroutermost LIA positions without a major readvance proba-bly because the drop in summer temperatures was accompa-nied by low winter precipitation (Bickerton and Matthews1993) In Jotunheimen and especially at the most continental







41 Study material
















42 Mapping LIA area


421 Landsat image






422 Aerial photos


























5 Results





Fig 6














6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References
























































































41 Study material
















42 Mapping LIA area


421 Landsat image






422 Aerial photos


























5 Results





Fig 6














6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References


























































































42 Mapping LIA area


421 Landsat image






422 Aerial photos


























5 Results





Fig 6














6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References





















































































422 Aerial photos


























5 Results





Fig 6














6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References




































































































5 Results





Fig 6














6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References



























































































6 Discussion



Fig 7








Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References





















































































Fig 9




Fig 10



















7 Conclusions





Edited by J O Hagen

References
































































































7 Conclusions





Edited by J O Hagen

References



















































































mapping glaciers in jotunheimen, south-norway, during the ...elevation model. “little ice age”...

Documents