Introduction

Deep-seated gravitational slope deformations (DGSDs) have been described and investigated in regional and case studies throughout the Eastern Alps (Ampferer 1939; Stini 1941; Clar and Weiss 1965; Zischinsky 1969b; Moser 1994; Crosta 1996; Hermann and Becker 2000; Reitner 2000; Weidner 2000; Agliardi et al. 2001), and they most commonly occur along deep glacial valleys in foliated rocks. Their initiation is thought to have taken place after the Late to the Postglacial retreat of the alpine valley glaciers that left behind instable over-steepened slopes (Reitner et al. 1993) and are usually accompanied with the gravitational reactivation of preexisting tectonic discontinuities. DGSDs commonly affect whole mountain slopes and have a typical morphology resulting from the gravitational sagging and spreading of the rock masses.

The deformation mechanism of DGSDs is commonly referred to as creep (Radbruch-Hall 1978; Hutchinson 1988) and is characterized by slow to extremely slow deformation rates (Cruden and Varnes 1996). This deformation process is often combined with partial sliding and stick–slip processes (Moser and Glumac 1983; Weidner 2000) resulting in a typical discontinuous deformation behaviour with deformation rates ranging between 0.3 and 700 cm/year (Moser 1996). The existence of continuous penetrating basal shear zones in DGSDs is still a matter of ongoing scientific debate (Noverraz 1996). Recent studies mainly focus on their detection by using geophysical methods (Ferrucci et al. 2000; Brückl and Brückl 2006); however, these investigations commonly do not extend over the whole length of the DGSD.

DGSDs are a common feature in the Tuxer Alps southeast of Innsbruck (Gruber 2005). Along the eastern slopes of the Watten Valley, a 20-km long, north–south striking valley branching off the Inn Valley, several DGSDs were detected (Madritsch 2004) that not only negatively affect human infrastructure but also contain aquifers for a number of springs that constitute the local water supplies (Fig. 1). The latter circumstance and the fact that DGSDs in the wider region are rarely well documented or have been investigated mainly from a geologic–geomorphologic point of view (Poscher 1990; Grasbon 2001; Gruber 2005) encouraged a hydrogeologic approach. The DGSD Sagspitz was chosen as a case site for the present study and revealed the usefulness of hydrogeological methods in the examination of the subsurface structures in DGSDs.

Fig. 1
figure 1

Geologic setting of the study area (framed by yellow square): Tectonic units (modified after Brandner 1980): 1; Central Gneiss, Penninic, 2; Bündner Schist, Penninic, 3; Innsbruck Quarztphyllite Complex, Lower Austroalpine, 4; Patscherkofel crystalline, Lower Austroalpine, 5; Tarntal Mesozoic, Lower Austroalpine, 6; Ötztal crystalline, Middle Austroalpine (after Tollmann 1977), 7; Brenner Mesozoic, Middle Austroalpine (after Tollmann 1977), 8; Kellerjoch Gneiss, Upper Austroalpine, 9; Northern Calcareous Alps, Upper Austroalpine, 10; Quaternary. Also shown are DGSD affected areas along the eastern slope of the Watten Valley as mapped by Madritsch (2004), as well as the locations of the rain gauges Kleinvolderberg and Wattener Lizum and of the springs s6 and s15 that were used to calculate the local altitude effect

Full size image

Geologic setting

The study area lies within the Lower Austroalpine Innsbruck Quartzphyllite Complex (Fig. 1). It is a monotonous succession of low-grade metamorphic Paleozoic sediments that was long treated as an undifferentiated rock mass until a rough division was established by Haditsch and Mostler (1982) and later enhanced by Rockenschaub et al. (2003).

The Innsbruck Quarzpyllite Complex comprises the lowermost Ordovician quartzphyllite–greenschist series, the calcareous and dolomitic marble bearing a carbonate–sericite–phyllite series and the youngest blackschist–carbonate series of the Upper Silurian to Devonian age, including graphitic schists and calcareous marbles (Fig. 2). Parts of the Innsbruck Quartzphyllite Complex in the study area consist of mylonitic quartzites and garnet-bearing mica schists. These rocks are supposed to be part of a diaphtoritic higher metamorphic zone of the Innsbruck Quarzphyllite Complex (Rockenschaub et al. 2003).

Fig. 2
figure 2

Lithostratigraphic chart of the Innsbruck Quartzphyllite Complex (modified after Haditsch and Mostler 1982 and Rockenschaub et al. 2003). Note the intercalations of gneisses, greenschists, marbles, and blackschists in the phyllitic rock mass. (1 sericite–quartzphyllites, 2 greenschists, partly calcareous, 3 porphyric and arenitic gneisses, 4 calcareous and dolomitic marbles, 5 calcareous black schists)

Full size image

The Innsbruck Quartzphyllite Complex has witnessed the Variscian and Alpine orogenies. It, therefore, shows an intense and polyphase ductile deformation resulting in a penetrating foliation and several fold generations (Kolenprat et al. 1999). In contrast to the southern part of the Innsbruck Quartzphyllite Complex, an overprinting axial plane foliation in the study area can only be observed in thin sections.

The brittle deformation of the Innsbruck Quartzphyllite Complex probably did not onset before Miocene times as indicated by its thermo-tectonic history (Fügenschuh et al. 1997) and was mainly controlled by the bordering major Alpine faults, namely the Inn Valley Shear Zone (Ortner 2003) to the north, the Northern Tauern Margin Fault System (Decker et al. 2003) to the south, and the Brenner Normal Fault (Selverstone 1988) to the west (Fig. 1).

The studied slope consists mainly of quartz– and sericite–phyllites (Fig. 3). Within this rock mass, intercalations of greenschist, gneiss, calcareous, and dolomitic marble are present (Fig. 4a). The intercalations are 1–10 m thick and cannot be traced for distances of more than 50–100 m due to ductile boudinage. White, coarse crystalline, calcareous marbles can be mapped to a larger extent at the scarp of the mass movement where they form the Sagspitz summit (2,410 m a.s.l.).

Fig. 3
figure 3

Geomorphic–hydrogeologic map of the DGSD Sagspitz (for location, refer to Fig. 1). The DGSD comprises the whole slope. The main scarp of the mass movement is located at the ridge of the slope where marbles and greenschists are intercalated in the phyllitic rock mass. Note that the rock is widely fractured to completely loosened at surface especially at the toe of the slope were it has a debris like character. Also shown are the monitored springs with their according signatures (see Table 1) and the spring district allocation (coordinates given in UTM32N). Springs s6 and s15 do not directly emerge from the DGSD Sagspitz but are located 500 m to the south of s4 and 4 km to the south of s13, respectively (see Fig. 1)

Full size image
Fig. 4
figure 4

Lithological properties of the Innsbruck Quartzphyllite Complex: a intercalations of competent greenschists in the incompetent phyllitic rock mass and b thin section of a typical quartzphyllite illustrating the alternation of quartz and muscovite rich layers that forms the penetrative foliation

Full size image

The phyllites are characterized by a penetrative foliation containing intrafolial isoclinal folds. Thin sections show that the foliation is formed by layers of phyllosilicates (muscovite, chlorite) and quartz (Fig. 4b). These phyllites are rheologically incompetent rocks (Fürlinger 1972) with wide joint spacing. The intercalations of greenschists, marbles, and gneisses react more competently and show more closely spaced brittle structures; in these rocks, the foliation is poorly developed.

The penetrative foliation of the phyllites dips gently towards NW, nearly parallel to the slope (Fig. 5a), and is assumed to form the main sliding surfaces within the rock mass.

Fig. 5
figure 5

Structural data: a stereographic representation of foliation measurements that shows a consistent gentle dip of the foliation towards NW (equal area lower, hemisphere projection, contoured pi-pole plot with pi-poles displayed as crosses) and b rose diagram showing that the majority of joints throughout the investigated area strike NNW–SSE and are steeply dipping (bars show strike directions at azimuth intervals of 20°)

Full size image

The most commonly observed brittle structures in the study area are E–W to NE–SW, striking sinistral strike–slip faults that are related to the Inn Valley Shear Zone (Ortner 2003) and the Northern Tauern Margin Fault System (Decker et al. 2003). N–S striking normal faults are related to the Sill Valley Fault, which is a brittle continuation of the ductile Brenner Normal Fault (Fügenschuh et al. 1997). These faults that strike parallel to the Watten Valley and the slope ridge were gravitationally reactivated during mass movement activity and represent the present day major joint and fracture orientation (Fig. 5b).

Materials and methods

The eastern slope of the Watten Valley was geologically and geomorphologically mapped at a scale of 1:10,000 with special emphasis being given to the structural geology and signs of relict and active gravitational movements. In addition, aerial photographs (Forest Inventory Survey recorded in 1953; approximate scale of 1:20,000; source: Regional Government Tyrol) as well as orthophotographs (recorded in 2000 at a scale of 1:5,000; source: Regional Government Tyrol; elevation model: BEV) were examined. Among several DGSDs detected along the eastern slopes of the valley (Madritsch 2004), the DGSD Sagspitz was chosen for detailed hydrogeologic investigation.

Out of over 150 mapped springs along the valley slope, 13 springs emerging from the DGSD Sagspitz were selected to be monitored (electrical conductivity, temperature, discharge) intensively for a time period of 1 year at monthly intervals. Springs s6 and s15 do not directly emerge from the mass movement but were also monitored for use as reference springs for the mean recharge altitude calculation of the spring waters. Water samples for δ18O analysis were taken monthly, and every 3 months, a chemical analysis of the spring waters was carried out. A δ18O analysis was also carried out on samples of precipitation obtained monthly from two gauges at Wattener Lizum and Kleinvolderberg (Fig. 1).

The electrical conductivity (reference, 25°C) and water temperature were measured in the field using a multifunctional probe (WTW LF 330). Discharge measurements were done manually using the stop watch and bucket method. Depending on the discharge rate, 1-, 3-, 10-, or 20-l buckets were used. Most major and minor chemical constituents of the water samples were measured using an ion chromatography system (Dionex DX 500 with autosampler AS40). NH4 and NO2 were determined by a flow injection system (Skalar SA4000 with Autosampler SA1050). Fe and Mn measurements were done with a Perkin Elmer Optima 3000 ICP. The analytical accuracies of these methods ranged from 2 to 5%. A Finnigan Deltaplus XL mass spectrometer was used for the δ18O analysis, and the values are reported in δ‰ units versus the Vienna Standard Mean Ocean Water (VSMOW) accuracies being ±0.1‰.

Mass movement characterisation

The DGSD Sagspitz, covering an area of approximately 3 km2, encompasses the entire slope from the ridge (Sagspitz summit at 2,410 m a.s.l.) to the valley floor (Fig. 3). The difference in elevation from the main scarp to the toe of the slope sums up to 1,100 m.

The DGSD Sagspitz is characterized by a geomorphologic appearance typical for mass movements of its kind (Hutchinson 1988; Moser 1994; Crosta 1996; Weidner 2000; Agliardi et al. 2001) and can be roughly divided into three major zones: (1) the double crested ridge zone (“Bergzerreissung”; Ampferer 1939; Zischinsky 1969b) in the uppermost part comprising the main scarp; (2) the sagging zone (Zischinsky 1969a) in the central part; (3) and the toe of slope (Bergzerreissung; Clar and Weiss 1965; Zischinsky 1969b) in the lowermost part.

Different types of gravitational rock deformation (Poisel 1998) are predominant in these zones, respectively. At the top of the mass movement, a double-crested ridge with wide N–S striking tension joints is developed (Fig. 6a). Here, toppling processes predominate over sagging. The sagging zone below is characterized by several generations of smaller scale scarps and counterscarps (Fig. 6b) and shows a rather gentle dip of slope (10–20°). In this zone, the rock is locally intensively fractured. The toe of the slope is steep (inclination, >30°) and affected by secondary shallow landslides. The rock mass at the surface is strongly fractured to completely loosened (development of debris-like deposits). Moser and Glumac (1983) suggested that a sliding component in this part of a DGSD is most likely. In profile view, the DGSD Sagspitz reveals a concave morphology in its upper part and a convex morphology in its lower part which is typical for DGSDs at a progressed stage of development (Zischinsky 1969a,b; Hermann and Becker 2000).

Fig. 6
figure 6

Morphologic features of the DGSD Sagspitz: a double-crested ridge at the top of the slope (2,400 m a.s.l.) featuring large and wide open tension joints. People ringed for scale. b Scarps and counterscarps in the sagging zone at around 1,700 m a.s.l. Both morpho-structural features function as potential infiltration zones

Full size image

The glacial erosion of the Watten Valley, the resulting high relief energy, and the pre-existing structural features in the Innsbruck Quartzphyllite Complex (penetrative foliation of the phyllites dipping parallel to slope, brittle faults striking parallel to the valley) set the framework for the formation of the DGSD Sagspitz.

No direct age constraints are available to date the initiation of the DGSD Sagspitz. However, geomorphic observation throughout the Watten Valley (Madritsch 2004) and neighbouring areas within the Tuxer Alps (Gruber 2005) have shown that, at higher elevations of between 2,100 and 2,400 m a.s.l., moraines of Lateglacial age (Daun to Egesen stage, Patzelt 1983) overlay DGSD related scarps. Furthermore, relict rock glaciers that are supposed to have formed at least 11,000 years ago (Barsch 1997) were obviously fed by the main scarps of preexisting DGSDs. This points towards an onset of DGSD activity in the Tuxer Alps already during Late glacial times (before 11,000 years), most probably due to the retreat of alpine valley glaciers that left behind over steepened slopes (Reitner et al. 1993). During the Holocene, at least parts of the DGSDs (secondary sagging bodies) were reactivated as evidenced by open-tension cracks and the geomorphic overprint of late glacial moraines by scarps at lower elevations (below 2,100 m a.s.l.). The above-listed observations apply to the DGSD Sagspitz as well. Its recent activity is restricted to secondary sagging bodies that were mapped at the toe of slope and in the sagging zone.

Spring hydrogeology

Field measurements

The monitored springs show mean discharges between 0.2 l/s (s14) and 12 l/s (s9). The mean water temperature lies between 3.5°C (s4) and 6.1°C (s15), and the mean electrical conductivity varies between 15 μS/cm (s6) and 232 μS/cm (s3; Table 1). These values are typical for springs emerging from the Innsbruck Quartzphyllite Complex (Burger et al. 2003; Millen et al. 2003).

Table 1 Mean, minimum, and maximum values of the field parameters of the monitored springs
Full size table

All field parameters exhibit annual variations with the temperature fluctuations of most springs generally following those of the air temperature. However, s1 hardly has temperature fluctuations indicating substantial mixing and/or retention in the aquifer and a deeper flow system. Discharge in all springs rises abruptly during spring and early summer. This increase is generally accompanied with a decrease in electrical conductivity (Fig. 7) and is interpreted as the effect of snowmelt in the aquifer systems. Springs s4, s5, and s6 discharge periodically (falling dry during the winter months), an indication that the aquifers are small and shallow, and/or freeze.

Fig. 7
figure 7

Spring discharge versus electrical conductivity. An increase in discharge is accompanied with an immediate to delayed drop in electrical conductivity. This drop is interpreted as the effect of snow melt infiltrating into the aquifer systems

Full size image

The electrical conductivity varies disproportionably with discharge and corresponds with the discharge altitude of the springs (Fig. 8). Differences in mineralization are due to retention times and/or local variations in aquifer lithology. Therefore, the electrical conductivity alone cannot be used for interpreting the size of the aquifer system.

Fig. 8
figure 8

Linear relation trend of electrical conductivity and discharge altitude. Deviations are probably due to local variations in aquifer lithology and hence water chemistry and/or longer retention times

Full size image

Hydrochemistry

Springs emerging from the Innsbruck Quartzphyllite Complex generally show a low degree of mineralization (Burger et al. 2003; Millen et al. 2003), and the water samples analysed in this study do not deviate from this observation (Tables 1 and 2, Fig. 8).

Table 2 Mean value of major anions and cations of water samples in mg/l (n = number of samples)
Full size table

Most of the observed springs can be classified as Ca–HCO3 and/or Ca–Mg–HCO3 water types (major cations and anions in descending order where %meq/l concentrations are >20%), the exception being springs s1, s2, and s3 which have a Ca–Mg–HCO3–SO4 type. Springs s1, s2, and s3 emerge at the toe of the slope and also have the highest mineralization.

The probable sources of the SO4 concentrations are sulphide minerals (e.g. pyrite), which are common throughout all the series of the Innsbruck Quartzphyllite Complex (Mostler et al. 1982). Sulphate minerals have not been recorded in the study area and there is no 1:1 molar correlation between dissolved Ca and SO4; this indicates that sulphide oxidation is dominant in the aquifer systems (e.g. Kilchmann et al. 2004).

The Ca, Mg, and HCO3 constituents of the spring waters are indicative of the dissolution of carbonate minerals promoted by CO2 uptake, which occurs commonly in the aquifers throughout the study area in the form of calcareous and dolomitic marble bodies and vein fillings. The very low contents of Na and K imply that substantial hydrolysis of the ubiquitous silicate phases (e.g. feldspars, micas) is not occurring.

δ18O analysis

The analysis of δ18O is commonly used to investigate the mean recharge areas of springs and has often been applied in alpine environments. Three methods to determine mean recharge altitude exist: (1) based on δ18O values obtained from precipitation gauges (pluviometers) at varying altitudes; (2) based on δ18O values obtained from “reference” springs with well-defined recharge areas at varying altitudes; (3) combination of methods 1 and 2 (Vogel et al. 1975; Rank et al. 1982; Stichler and Zojer 1986; Schotterer et al. 2000; Millen 2001; Yehdegho and Reichl 2002).

The method makes use of the fact that during the transformation of atmospheric water vapour to precipitation, which then infiltrates and becomes groundwater, fractionation processes between the heavier 18O and the lighter 16O isotope within the water molecules occur (Dansgaard 1964; Ehalt and Knott 1965). During condensation, the heavier 18O is enriched relative to the 16O in precipitation whilst becoming depleted in the remaining vapour.

It has been observed that the isotopic composition of precipitation changes with the altitude of the terrain and becomes more and more depleted in 18O at higher altitudes. This relation is known as the altitude effect (for review, see Mook 2000). The altitude effect is temperature related and is, furthermore, influenced by the rainout effect. Rain clouds that continuously collide with mountain chains and condensate are continuously depleted in the heavier 18O. Therefore, rains falling at higher altitudes further up the mountain are derived from an already 18O-depleted source and have lower δ18O values. Depending on the precipitation source, transport distances, and the local topography, i.e. weather divides, δ18O values vary greatly throughout the Alps (Humer et al. 1995; Schotterer et al. 2000; Kaiser et al. 2002; Kralik et al. 2003). These circumstances require the calculation of the local altitude effect within the survey area.

In this study, the calculation of the altitude effect was carried out firstly using only reference springs. In a second step, two precipitation gauges were considered as well. The precipitation gauge Kleinvolderberg (660 m a.s.l.) lies 9 km northwest of the study area within the Inn Valley; the gauge Wattener Lizum (1970 m a.s.l.) is 7 km to the south (Fig. 1). δ18O values have been derived from monthly mixed samples of precipitation and have been weighted against the amount of monthly rainfall to obtain the annual mean weighted δ18O value (Table 3). Note the maximum of the δ18O values during summer and, accordingly, the minimum δ18O values during winter; this is known as the seasonal effect (Epstein 1956; Dansgaard 1964).

Table 3 Results of δ18O analysis of the monthly precipitation from the Kleinvolderberg and Wattener Lizum precipitation gauges (location of the gauges is given in UTM32N coordinates)
Full size table

For the reference spring method, the selected springs need to have a well-definable small recharge area, determined from geomorphologic and hydrogeologic mapping, hydrochemistry, and the spring characteristics (e.g. low discharge and mineralization that indicate a small local aquifer) so that their mean recharge altitude can be appropriately estimated (Millen 2001; Yehdegho and Reichl 2002). Springs s5, s6, and s15 (Tables 1, 2, and 4) were selected as references, as they fitted the above criteria well.

Table 4 Results of δ18O measurements (mean, maximum, and minimum values) and mean recharge altitude calculation based on mean δ18O values from the reference springs only (0.18‰‰/100 m) and combined method using reference springs and the precipitation gauge Kleinvolderberg (0.24‰/100 m; n = number of measurements)
Full size table

The mean δ18O values and determined recharge altitudes of the reference springs were then used to construct a regression line revealing the relationship between δ18O value and altitude (Fig. 9). The resulting gradient of 0.18‰/100 m corresponds well with typical values from the Alps determined from precipitation or combined methods (Schotterer et al. 2000; Kralik et al. 2003).

Fig. 9
figure 9

δ18O regression line showing the calculated local altitude effect (0.18‰/100 m) using reference springs only: The closed squares indicate the mean δ18O value versus the discharge altitude of the springs. The open squares (s5r, s6r, and s15r) represent the determined mean recharge altitude and mean δ18O value of the reference springs. Uncertainties concerning the mean recharge altitudes are shown by the error bars. The mean recharge altitude can be determined by projecting the squares horizontally onto the regression line and then vertically onto the altitude axis. Furthermore, the horizontal distance of the springs to the regression line gives an indication of the size of the aquifer and depth of flow

Full size image

In a second step, the weighted mean δ18O values of the precipitation were plotted against the altitude of the gauges into Fig. 9. A second regression line was then constructed considering the values obtained from the Kleinvolderberg gauge (Fig. 10) and the well defined reference springs s5 and s6, resulting in a slightly higher local altitude effect of 0.24‰/100 m. To be noted here is the totally erroneous position of the Wattener Lizum gauge in this plot (Fig. 10). The high δ18O value could be due to a problem with correct snow amount measurements in winter (less snow registered than actually fallen) or due to the influence of Mediterranean precipitation in the area of this collector that is generally characterized by higher values than those derived from the Northern Atlantic and the North Sea (Kaiser et al. 2002; Kralik et al. 2003). These circumstances make the Wattener Lizum gauge inappropriate for recharge altitude calculation purposes.

Fig. 10
figure 10

δ18O regression lines showing the calculated local altitude effect using reference springs (0.18‰/100 m) and the precipitation gauge Kleinvolderberg (0.24‰/100 m) compared to Fig. 9. Note the totally erroneous position that the Wattener Lizum gauge has in this plot making it unsuitable for use in the recharge altitude calculations (see text for further discussion)

Full size image

The mean recharge altitudes of the springs (Table 4) were determined by extrapolating their discharge altitude onto the regression lines in Figs. 9 and 10. The calculated error of ±150 m for this method is based on the uncertainty of the assumed mean recharge altitude of the reference springs that is in agreement with the study of Yehdegho and Reichl (2002).

Furthermore, the horizontal distance to the regression line gives a good indication about the relative size of the spring aquifer and depth of flow, i.e. the further away the spring plots to the left of the regression line, the larger the aquifer and greater depth of flow. This implies that the springs s1, s2, and s3 have large aquifers and greater depth of flow in comparison to s9, s10, and s11.

The annual trend of the δ18O values obtained from spring water can also be used to reveal information on the retention potential of an aquifer (Stichler and Hermann 1983). Spring waters with longer retention times have a δ18O trend with little variation, whereas those with short retention times show larger fluctuations. In this case, the reference springs s5, s6, and s15 have a shorter retention time than the rest of the springs, also an indirect indication of their small local aquifers (Table 4; see deviation from mean).

Snowmelt water plays an important role in the recharging of the aquifers in the study area. It has a significantly lower δ18O value than summer precipitation and can, therefore, be used as a natural tracer. Figure 11 shows that a decrease in δ18O is mirrored by an increase in discharge, which can be related to snowmelt water infiltration into the aquifer system. This further shows that some aquifers have rather poor retention properties, allowing little time for groundwater mixing to occur. Such springs also generally have a lower mineralization. Some of the springs (s1, s2, and s3), however, do not show this relation. In fact, the δ18O value increases with the rising discharge. At these springs, the increased discharge does not consist of snowmelt but of well-mixed groundwater were the δ18O signature has been diffused due to a longer retention time. These springs also have the most chemically developed water in the study area (Table 2).

Fig. 11
figure 11

Relationship between δ18O value and discharge trend: δ18O of s1 increases with discharge increase; δ18O of s8 decreases with discharge increase (refer to text for explanation)

Full size image

Differentiation of spring districts

Based on geomorphologic and hydrogeologic mapping, field parameter trends, hydrochemistry, and δ18O values, four spring districts can be distinguished within the DGSD Sagspitz (Fig. 3).

Springs of district 1 (s5, s6) are located at an altitude of approximately 2,100 m a.s.l. and are characterized by a periodic discharge with a maximum during spring and zero discharge during winter. δ18O values (maximum of −12.82‰ to a minimum of −13.94‰VSMOW), electrical conductivity (minimum of 8 μS/cm to a maximum 56 μS/cm), as well as water temperature show large annual variations, and their waters are of low mineralization. Such factors are indicative of short retention times and small shallow aquifers. The springs emerge from talus, completely disintegrated quartzphyllite due to sagging, or from glacial sediments.

Springs of district 2 (s4, s7, s8) emerge between 1,700–1,900 m a.s.l. They have a continuous discharge and low electrical conductivity (minimum of 50 μS/cm to a maximum of 94 μS/cm). The results of the δ18O analysis (maximum of −13.11‰ to a minimum of −13.59‰VSMOW) indicate that in comparison to the springs of district 1, these springs have larger aquifers and deeper flow paths, i.e. the vertical difference between the calculated mean recharge area and discharge area lies between 400 and 600 m. Springs of similar characteristics can also be found at the toe of a slope (s14).

Spring district 3 (s9, s10, s11) encompasses springs with a more stable annual discharge trend that is less affected by precipitation events. These springs have a higher mean mineralization (102 to 174 μS/cm) in comparison to districts 1 and 2. The vertical difference between the calculated mean recharge area and discharge area lies around 600 m.

Springs of district 4 (s1, s2, s3, s13) all emerge at the toe of the mass movement. They have a relatively high and stable mean water temperature (4.4 to 5.1°C); the mean flow rates are relatively large (5.5 to 7.9 l/s), and the mean electrical conductivity is >144 μS/cm. Especially springs 1 to 3 have the highest mean values of electrical conductivity (>210 μS/cm) and have comparatively well-developed waters (Ca–Mg–HCO3–SO4 water type) which are an indication of longer retention times. The springs of district 4 also have the most stable δ18O values (Table 4; see deviation from mean) and water temperature trends, this implying a large and well mixed source. Most important, however, are the results of the δ18O calculations that indicate that these springs have the highest mean recharge altitude and greatest vertical distance between the recharge and discharge areas (800–1,000 m). Further, the recharge area corresponds with the double crested ridge area of the DGSD that contains many outcrops of highly fractured marbles—a perfect lithology for allowing rapid and deep(?) infiltration of precipitation into the DGSD.

Conclusions

The Innsbruck Quartzphyllite Complex within the area of the DGSD Sagspitz contains a large number of small- to medium-sized springs (mean values, 0.2 to 12 l/s). The aquifer systems can be directly related to the deep-seated slope deformation that caused intense fracturing of the rock and the development of secondary permeability and porosity.

The hydrochemistry and calculated mean recharge altitudes of springs (δ18O analysis) emerging from the Sagspitz DGSD imply that the waters emerging at the toe of the slope (district 4) infiltrate at the main scarp of the mass movement. This observation supports the hypothesis of the existence of continuous flow paths from top to bottom through the DGSD. Snowmelt waters are not registered in the springs of district 4 but are mixed within the aquifer before emergence, indicating good retention properties for the aquifers of this spring district. Furthermore, the annual temperature and δ18O variations throughout the hydrological year show that these springs are not influenced by run-off infiltration from other spring districts. Interpretations concerning the internal structure of the DGSD based on the hydrogeologic data are shown in Fig. 12. The aquifer boundaries are related to shear zones in the DGSD. In the Innsbruck Quartzphyllite Complex, such shear zones, to a large extent, consist of clay gouge-bounded kakirite and are underlain by more compact and less disturbed rock. Therefore, these horizons act as significant aquitards. Water infiltration and flow occurs in the overlying loosened and fractured rock mass.

Fig. 12
figure 12

Cross-section through the DGSD Sagspitz based on the geomorphologic, hydrochemical, hydrological, and δ18O analysis. The calculated local altitude effect on the δ18O values is displayed on the left axis. Derived spring districts are shown with representative mean δ18O emergence values. Arrows indicate runoff infiltration and groundwater flow paths. The latter are related to the shear zones within the deep-seated slope deformation (shaded area). See text for further discussion

Full size image

In the spring district 4 aquifer systems, the aquitard is interpreted to be the basal shear zone of the DGSD with continuous groundwater flow occurring above this horizon through the mass movement from the uppermost scarp to the toe of the slope. Spring districts 2 and 3 are more local and related to secondary sagging bodies and shear zones within the DGSD. Therefore, a multilevel aquifer system is very apparent within the DGSD.

The conceptual model of the DGSD Sagspitz results in a maximum vertical mass movement thickness of 150–200 m, which is well in agreement with results from other case studies that applied drilling (Noverraz 1996) and geophysical prospecting methods (Ferrucci et al. 2000; Brückl and Brückl 2006).

Holocene mass movement activity might be restricted to the secondary sagging bodies in the central part of the DGSD and to superficial landslide activity at the toe of slope.

The current study shows that the use of isotope analysis on spring water, in addition to hydrogeologic (Pergher and Burger 2004) and geomorphologic standard investigations, is a useful tool when investigating the properties and structures of DGSDs.