Okshola1

8/4/2019 Okshola1

1/17

123

Morphology and speleogenesis of Okshola,

Fauske, northern Norway: example of a multi-stagenetwork cave in a glacial landscape

Rannveig vrevik Skoglund & Stein-Erik Lauritzen

Skoglund, R.. & Lauritzen, S.E.: Morphology and speleogenesis of Okshola (Fauske, northern Norway): example of amulti-stage network cave in a glacial landscape. Norwegian Journal of Geology, Vol. 90, pp 123-139. Trondheim 2010.ISSN 029-196X

Maze, or labyrinth, caves are high-porosity zones in karst. Reticular networks may arise through different speleogenetic processes. Here, we presentand discuss an apparently multi-stage labyrinthal development in a stripe karst setting in the Norwegian Caledonides. Okshola (the upper part ofthe Okshola-Kristihola cave system at Fauske, Nordland) displays a network of preserved, essentially phreatic tubes intersected by four distinct,

vadose inlet passages. The cave developed along a low-angle fracture (thrust) zone, which is sub-parallel with the foliation. Scallops in the walls ofphreatic conduits demonstrate that water flow was directed down-dip into the rock mass, and thus that the phreatic network developed during thelast active stage as a groundwater recharge zone. This flow function is consistent with the proximal location of the cave with respect to former topo-graphically directed glacial flow. Cyclic and strong fluctuations in the hydraulic regime are evident from cave interior deposits. We suggest that Oks-hola developed in concert with the glacial erosion of the surface topography and that a process of caprock stripping resulted in progressive loweringof both sink and spring levels. Morphology, together with radiometric datings, indicate that speleogenesis commenced several glacial cycles ago.

Rannveig vrevik Skoglund, Present address: Department of Geography, University of Bergen, Fosswinckelsgt. 6, 5007 Bergen, Norway. E-mail: [email protected]. Stein-Erik Lauritzen, Department of Earth Science, University of Bergen, Allgaten 41, 5007 Bergen, Norway; Stein-ErikLauritzen, Department of Plant and Environmental Sciences, The Norwegian University of Life Sciences, 1532 s, Norway.

IntroductionThe study of limestone caves in relation to their geomor-phic and geological setting provides opportunities forinvestigating landscape evolution. Whereas at the surface,landscapes may become obliterated by erosion, evidenceof past conditions can be preserved underground. Here,we present recent results of detailed mapping and ana-lysis of a major part of the Okshola-Kristihola cave systemat Fauske, Nordland. The cave is the second longest cavesystem in Norway, situated beneath the marginal slope ofa glacial trough near Fauske. Recent mapping (this work)and estimates of yet unmapped passages suggest an aggre-gate length exceeding 12 km, comprising two complex net-work- or labyrinth zones. These are intersected and linkedby several deeply incised streamways, indicating a corre-spondingly complex, multi-stage development over rela-tively long time-spans.

The geological setting of the Okshola cave system belongsto a type known as stripe karst that occurs throughout theNorwegian Caledonides and elsewhere (Lauritzen 2001).Caves are developed in relatively thin marble layers whichare bounded stratigraphically by schists with low permea-bility, the whole sequence often being intensely deformed

by folding and subsequently fractured. In Norway stripekarsts almost always occur in landscapes that are heavilysculptured by glacial erosion. Glacial unloading has played

a part in forming or dilating fractures parallel to the landsurface, along which the caves developed. Thus Oksholaprovides an example in which glacial erosion and under-ground speleogenesis are closely related. Okshola itself is ofparticular interest in this connection because it comprisesa series of labyrinths of phreatic tubes, intersected by morenormal linear passages. These labyrinths (also known asmaze caves) represent an extreme development of poros-ity by karstic dissolution, a phenomenon that is importantin hydrogeology and for water supplies as well as for res-ervoir development in oil-bearing carbonate strata. TheOkshola labyrinths occur in a different setting comparedto the classic descriptions of maze caves where the empha-sis has been on the role of recharge through non-carbonatestrata. The Okshola labyrinths are clearly related to lateralrecharge via the marble outcrop, so the circumstances oftheir origin must differ significantly from those surround-ing classical mazes.

Previous work

A considerable number of karst caves within meta-carbonates in the Scandinavian Caledonides comprisereticular networks, where groundwater flow occurred

in a flow net rather than through linear conduits. Twostructural situations seem to promote network (labyrinth)architecture: steeply dipping strata (greater than about

NORWEGIAN JOURNAL OF GEOLOGY Morphology and speleogenesis of Okshola

8/4/2019 Okshola1

2/17

124

60) and low-dip strata (lower than about 30) (Lauritzen2001). Steeply dipping strata yield tiered networks, whilelow-dip strata produce correspondingly gently dippinglabyrinths. Several of the longest caves in Norway are lab-yrinths.

Regional and contact metamorphism during the Cale-donian orogeny produced crystalline carbonates (i.e.marbles) in which all previously existing voids wereeliminated. This resulted in an almost impermeablerock matrix that can be likened to granite or other mas-sive crystalline rocks. Water circulation (and hence spe-leogenesis) is therefore guided by tectonic and unload-ing fractures produced in later brittle regimes (Lauritzen1989a, 1991a, 2001). These regimes are linked to the longsequence of post-Caledonian plate-tectonic events thataffected Norway, including Tertiary uplift and glacial ero-sion and unloading (e.g. Talwani & Eldholm 1977; Japsen& Chalmers 2000). In this setting, the formation of the

guiding fractures (i.e. those fractures that developed intocave conduits) pre-date subsequent cave development andthe timing of the corresponding tectonic events therebyprovides an upper boundary for the potential age of agiven cave passage. Apart from terrain-parallel unload-ing fractures - which are linked to events of erosion andglacial unloading (Harland 1956; Lauritzen 1986) - mostfracture zones in Norway existed long before Quater-nary speleogenesis commenced (e.g. Gabrielsen et al.2010). No evidence of pre-glacial interior cave deposits orhypogene (hydrothermal) speleogenesis has been foundanywhere in northern Norway. So far, this may be taken

as supportive (i.e. negative) evidence that the presentkarstification is of Quaternary age.

Radiometric dating of cave deposits in Scandinavia hasdemonstrated that speleogenesis commenced prior tothe limit of the U-series time-range (> 750 ka) (Lauritzen1991b, and unpublished dates), suggesting that several ofthe present caves might have commenced their develop-ment prior to the mid-Pleistocene transition (MPT) at800 ka (Ruddiman et al. 1989). Speleogenesis thereforebegan prior to, and continued through, most of the upperPleistocene during which time hydrological conditions varied periodically between subglacial and interglacialextremes.

Description of field area

Geomorphic and geological setting

The Okshola-Kristihola cave system (6715N 1530E) issituated on the northern side of the valley of Nedrevatn(here termed Nedrevatn valley), the eastern extension ofSaltfjord-Skjerstadfjorden. This E-W-trending fjord val-ley is glacially overdeepened and cuts the Caledonian

strike direction (Fig. 1). The cave system is situated inthe distal part of a small tributary valley with a presentdrainage area of approximately 10 km2.

The bedrock in the Fauske area comprises Caledonianthrust nappe complexes with overall NNE-SSW trendingfoliation and fold axes situated over older basement (Ste-phens et al. 1985). Bedrock in the field area belongs to theUppermost Allochthon (Rdingsfjll Nappe Complex)and consists mainly of marble and mica schist (Gustav-

son et al. 2004). These strata have undergone amphibo-lite facies metamorphism (Stephens et al. 1985). Foldingthrough several fold phases has given the rock sequencea complex fold interference pattern (Fig. 2). OksholaCave is developed in grey calcitic marble (Rognan group,Fauske Nappe) underlying layers of mica schist and con-glomerate (Gustavson et al. 2004). Stratigraphically thinbeds of marble exposed over long distances, and isolatedand constricted by aquiclude wallrocks are known asstripe karst(Horn 1937; Lauritzen 2001).

Mica schist and other wallrock may contain iron oxideores or iron and base metal sulphides, which impregnate

the marble at the contacts (Lauritzen 2001). Pyritic oxi-dation and subsequent sulphuric acid corrosion seemquite common in marble stripe karst in northern Nor-way and may be an important mechanism during earlystages of cave inception (Lauritzen 2001).

Glacial history

During the last glacial maximum, the Fennoscandianice sheet covered all of Scandinavia and the continentalshelf. The regional ice flow direction was towards theNW (Andersen 1975; Ottesen et al. 2005). During degla-

ciation, ice lobes in fjords and valleys drained the rem-nants of the ice sheet. Andersen (1975) reconstructed theregional history of this deglaciation. The Finneid ice lobe(which occupied the Nedrevatn valley), for example, wasfed from the ice sheet that covered the mountain districtsto the east (near the Swedish border) (Fig. 1a). The Salt-dalsfjord ice lobe was fed from the south and the moun-tain area to the east of it. During the Younger Dryas,these lobes (among others) joined in the main fjord areaand deposited a submarine end moraine at the mouthof Saltfjord-Skjrstadfjorden, more than 50 km distallyfrom the cave. During the Pre-Boreal, the Finneid icelobe deposited at least three distinct end moraines in theFauske area: ines-Holstad (H), Finneid (F) and vrevatn () (Fig. 1a). Lateral moraines deposited by theFinneid ice lobe are found between 400 and 600 m a.s.l.south of lake vrevatn. At the time, the sea level in Salt-dalsfjorden (10 km south of the cave) was at 130 m a.s.l.,while at the cave location, the upper post-glacial marinelimit (ML) was about 110 m a.s.l. (Fig. 1b). At present,there are two small mountain glaciers in the eastern partsof the drainage area of Nedrevatn valley.

Speleological setting

Okshola and Kristihola are both maze caves (Fig. 3a).The Okshola cave is the upper part of the system, andhas been re-surveyed over a length of about 8.4 km.

R. . Skoglund & S.-E. Lauritzen NORWEGIAN JOURNAL OF GEOLOGY

8/4/2019 Okshola1

3/17

125NORWEGIAN JOURNAL OF GEOLOGY Morphology and speleogenesis of Okshola

Fig. 1. a) Map of the Fauskearea. Blue arrows: glacier flowduring last deglaciation. Direc-tion of glacier flow, striationsand end moraines from Ander-sen (1975). Full name of endmoraines: see text. Contourinterval: 100 m. White rectan- gle: Area of Fig. 2. Inset: Keymap to the investigated area. b)Cross-section of the glacial inci-sion in the Nedrevatn valley, loo-king westward in the direction offormer glacier flow. Vertical axisis exaggerated 5 times. Bedrockdistribution from Gustavsonet al. (2004) with our interpre-tation of folding. ML (marinelimit) according to Andersen

(1975).

Fig. 2. Map of cave area with marble outcrop (blue)

(Gustavson et al. 2004), Okshola-Krisithola cave sys-tem (black) and present sink and spring locations (red).Contour interval: 100 m.

8/4/2019 Okshola1

4/17

126

The main entrance is a large arch at the bottom of thestream sink doline of the Okshola River, approximately160 m a.s.l. (Fig. 2). The river drains the tributary val-ley and flows 1.5 km SSE through the second part of thesystem, Kristihola, and emerges at Heitosen spring, atsea level. Okshola has been surveyed upwards to 260 ma.s.l., whereas the lowest accessible point of Okshola isthe siphon of North River at 140 m a.s.l. The lowermostelevated siphon in Kristihola is situated at about sea level(Heap 1969).

The active streamway passage of Kristihola (i.e. theStreamway) connects the two caves. It is a huge vadosetrench (up to 30 m high and 9 m wide) cut by vertical

erosion below a lens-shaped phreatic tube in the roof(Fig. 3b). The ancient streamway is blocked by collapsedrock so that the present connection between the twocaves is through narrow floodwater passages. The entireKristihola cave is situated at a lower elevation than theentrance to Okshola.

The structural guiding and hydrological function of thetwo network sections seem to be markedly different.From bedrock maps (Fig. 2) it is evident that the rocksequence exhibits a complex fold interference pattern.

This has resulted in foliation surfaces dipping in oppositedirections in Kristihola and Okshola (Fig. 3).


Fig. 3. a) Map of the Oks-hola-Kristihola cave system,based on present survey ofOkshola and previous surveyof Kristihola by the KendalCaving Club and U. Hol-bye (Heap 1969). Red lines:contours of floor elevation(m a.s.l.). Red dashed lines:extrapolated contours inaccordance with informationon the cave map of Kristi-hola. The dip direction ofthe cave plane (foliation and guiding plane) is perpendi-cular to contours. Note thatcontours only exist within passages. b) Cross-sectionsof the maze area in Oks-

hola: A-A and Kristihola:B-B. Scale: 3-times exag- geration in relation to cavemap. Note that cave cross-sections are oblique to dipdirection. c) Hypothetical fold interference pattern ofthe foliation and the guid-ing plane in accordance withdip directions observed inthe cave and obtained fromcontouring. The interference

pattern is a result of several(at least two) fold phases.

8/4/2019 Okshola1

5/17


Methods

The Okshola-Kristihola cave system was first surveyedby members of the Kendal Caving Club (Heap 1969)and later extended by U. Holbye (unpublished map-ping). In order to obtain an accurate and complete 3D

map, Okshola was completely resurveyed to BCRA grade5C (Day 2002), where detailed examination of morpho-logical features was emphasized. During our survey, caveinterior details such as sediments on the surface of floordeposits and speleothem distribution were recorded, andfractures and foliation logged. From survey data, a 3Dmodel of the cave was obtained in the Grottolfcave surveyprogram (Lauritzen 2002; Lauritzen 2004). This offered aunique opportunity to investigate structural and morpho-logical elements of the cave.

Geometrical parameters of the entire cave and cave sec-tions were calculated in the Grottolfprogram: Cave length

- total passage length. Cave depth - vertical differencebetween the highest and lowest passages. Cave area - hor-izontal plan area of the cave. Cave volume - total volumeof all cave passages based on elliptical cross-sections. Therock area that circumscribes the cave (or cave section)is estimated as a convex hull, i.e. an area of only convexangles enclosing all cave passages. The correspondingrock volume that contains all cave conduits is calculatedas the convex hullarea multiplied by the maximum pas-sage height. From these data other speleometric datacan be calculated: Passage cross-sectional area - cave vol-ume divided by cave length, Passage density - cave length

divided by convex hull area, Areal coverage - cave areadivided byconvex hullarea (%) and Cave porosity - cavevolume divided by rock volume (%).

The fractal dimension or box dimension, D, (in twodimensions) is a measure of how completely the cave fillsa plane of projection. It is determined by standard box-counting (Feder 1988). When the number of boxes cov-ering the filled outline of the cave is plotted against thelength of the box side in a log-log-diagram, D can bedetermined as the slope of the plot (Kusumayudha et al.2000). This approach contrasts with the volumetric, mod-ular method of Curl (1986), but serves our purpose bestbecause the cave is essentially 2-dimesional due to thestripe geometry.

Conduit morphology provides information about thehydrological conditions under which the conduitsevolved (e.g. Lauritzen & Lundberg 2000; Ford & Wil-liams 2007). Under water-filled (phreatic) conditions,corrosion acts in all directions thus forming conduitswith circular or elliptical cross-sections but is also depen-dent on the structural guiding and lithology. In air-filledconditions (vadose), with a stream along the passagefloor, corrosion and erosion act under the influence of

gravity forming a vadose trench or canyon giving theconduit a keyhole or T-shaped cross-section. If, underphreatic conditions, sediments cover the passage floor or

fill the entire passage, symmetrical corrosion is impededand further water flow may create channels and pocketsin the conduit ceiling. This process is termed paragen-esis (Renault 1968) or antigravitational erosion (Pasini2009). Breakdown morphology is characterized by jaggedsurfaces of rupture in walls and roof and piles of angular

rock accumulated along the passage floor (Ford & Wil-liams 2007).

Scallops are asymmetric flow marks in cave walls formedby aquatic corrosion. Following procedures described inLauritzen (1982) and in Lauritzen & Lundberg (2000),scallops in walls of phreatic conduits can be used todetermine the direction of paleowater flow and to make arough estimate of flow velocity.

The method of cave map contouring was first used byFord (1965). Contours are drawn (interpolated) betweenpassages with the same floor elevation and with a gentle

curvature along the shortest distance. Systematic patternsin the contour distribution may reveal the underlyinggeological control of the passages. Closer-spaced con-tours exhibit steepening of the gradient while dramaticcurvature shows elevated or lowered passages (in relationto the regular pattern).

A qualitative tracer experiment with optical bright-ener (Photine CU) was carried out as standard proce-dure (Glover 1972). An injection was made in the NorthRiver siphon and recorded on unbleached cotton in cavestreams, surface creeks and springs. Springs and flow

paths of the present drainage system were identified andgave information about possible inaccessible karst con-duits and paleo flow paths.

Results

Cave description

MorphologyWe have identified 8 distinct morphological areaswithin Okshola (Fig. 4). Four maze labyrinthal areas arecharacterized by conduits of circular, elliptical or len-ticular cross-sectional shape: Upper Maze (UM), Cen-tral Maze (CM), Lower Maze (LM) and Inner maze (IM)(white areas in Fig. 4). The mean passage cross-sectionalarea of these sections is markedly different: UM 2.9 m2,CM 3.6 m2, LM 6.7 m2 and IM 1.4 m2. The passage den-sity is highest in UM (141 km/km2). Speleometric datafor the entire Okshola and four cave sections are givenin Table 1.

The cross-sectional shapes of the maze conduits implysymmetrical dissolution radially away from the guidingstructure under water-filled conditions. These conduits

are preserved phreatic tubes with some minor modifi-cations. Small vadose incisions and paragenetic featuressuch as halftubes and dissolved fissures in the ceiling do

8/4/2019 Okshola1

6/17

128 R. . Skoglund & S.-E. Lauritzen NORWEGIAN JOURNAL OF GEOLOGY

Fig. 4. Left: Morp-hological sections ofOkshola. White areasare dominated by pre-served phreatic tubeswhile areas in differentgrey shades have suffe-red from modificationeither by vadose riversor breakdown. A-B:Cross-section in Fig. 8.Right: Typical passagecross-sections. Meanpassage cross-sectionalarea of each section is given in parenthesis.Scale is 5 times exaggerated in relation tothe cave map to the

left.

Table 1. Speleometric data for the entire Okshola cave and four sectionsOkshola Upper

Maze

Central

Maze

Lower

Maze

Central

FMRC-zone

Length, 103 m 8.4 1.3 2.1 0.9 1.6

Depth, m 123 29 58 46 60

Cave area, 103 m2 41.2 4.1 8.1 4.6 9.9

Cave volume, 103 m3 62.3 3.7 7.3 6.0 17.4

Convex hull, 103 m2 139.4 9.0 26.1 12.3 21.1

Maximum passage height in each section, m 18 4.5 3 7.5 10

Rock volume, 106 m3 2.5 0.04 0.08 0.09 0.21

Mean passage cross-sectional area, m2 7.4 2.9 3.6 6.7 10.7

Passage density, km/km2 61 141 79 72 77

Areal coverage, % 30 45 31 37 47

Cave porosity, % 2.5 10.2 9.4 6.9 8.2

Fractal dimension 1.6 1.6 1.6 1.7 1.5

8/4/2019 Okshola1

7/17

129

exist in various places (Fig. 4). Passages towards the sur-face (hillside) terminate quite frequently in collapse fea-tures.

The three outer labyrinthal areas are separated from eachother by two trunk passages with large elliptical cross-

sections and distinct secondary vadose openings: theUpper Entrance (UE) and Fata Morgana (FM). Thesetwo passages belong to cave sections characterized bydistinct modifications either by marked vadose canyonsor breakdown features. However, minor areas showingparagenetic features do occur in these sections as well.Four different sections were identified: North River(NR), UE, Fata Morgana-River Canyon zone (FMRC)and Entrance-Icehall (EI) (marked by different grey-shades in Fig. 4). The mean passage cross-sectional areaof the four sections differs and increases southwards(down-gradient) from 5 m2 in NR to 27 m2 in EI (Fig. 4).

Large parts of these modified sections have characteristickeyhole or T-shaped cross-sections, i.e. subcircularor lenticular phreatic conduits along the ceilings withsecondary incised vadose canyons in the floors (Fig.4). NR and UE consist of single passages except fromupstream schist horizons which have restricted down-cutting (Fig. 4). The FMRC-zone is distinguishedby large passages (

8/4/2019 Okshola1

8/17

130

water of North River disappears in a siphon at the end ofthe RC passage. A qualitative (i.e. point-to-point) tracingexperiment demonstrated that the flow diverges. Tracerbreakthroughs were detected both in the southwesternsurface stream (here abbreviated to SW Spring) and inthe cave stream in Kristihola, and subsequently in the

Heitosen spring.The vadose canyons have four distinct inlet passagesand flow paths marking previous open channel flowconditions (Fig. 6). Scallops in the walls and ceilings ofthe phreatic tubes (formed by corrosion when the con-duit was water-filled), display a consistent flow directioninto the cave, and away from the surface (Fig. 6). Thisdemonstrates that the phreatic network had an influentflow function. In Lower Okshola the flow pattern seemsto have diverged from the trunk passage, FM. The mostabundant scallop lengths vary between 10 and 20 cm.

These correspond to a roughly estimated water velocityof 15-30 cm/s (Lauritzen & Lundberg 2000).

The marbleThe cavernous marble contains numerous folded schistlenses and schist horizons. These aquicludes have so

restricted passage development that they occasionallyform the passage ceiling or floor. The interface betweenmarble and schist is commonly gradual. A zone ofrusty marble and rusty schist exists between the micaschist roof and the pure marble. The presence of pyrite(sulphide) within the rock is also revealed by numer-ous gypsum crusts and small drip-pits in the underlyingmarble formed by the acidic water.

Structural speleology

Guiding planeA single low-dip fracture zone is the dominant guiding

structure in the entire Okshola-Kristihola cave system.It is detectable in most cross-sections (Fig. 4) and hasapparently been favourable for water penetration.Broadly speaking, Okshola was initiated along thislow-dip guiding plane. (It is termed guiding plane todistinguish it from the steeply inclined guiding frac-tures.)

The appearance of the guiding plane varies between asingle fracture plane and lenticular zones of dense frac-turing (Fig. 7). The internal fracture pattern in this low-angle zone is interpreted as possible en echelon Riedel

shear planes with insignificant movement. This sug-gests that the guiding plane is a zone of incipient shear.The orientation of the interpreted R-shears indicates theprescence of a reverse-sense shear zone with the greatestprincipal stress,

1(i.e. compression), oriented NE-SW,

whereas the least principal stress, 3

(i.e. tension), wasvertical.

Few good measurements of the guiding plane wereobtained due to its low dip and tendency to form dis-solutional recesses. Accordingly, the orientation of theguiding plane in the various sections was determined bythe 3D model in the Grottolfprogram. In the 3D model,the cave is displayed as a single plane which is slightlycurved in an open, antiformal fashion (Fig. 8). This isconsistent with the pole plots of the guiding plane occur-ring along a great circle in the stereographic projection(Fig. 9a). The hinge line dips 13 towards the SW (226).Pole plots of the guiding plane and the foliation coin-cide in the stereographic projection, meaning that thesestructures are overall parallel (Fig. 9a). In some places theguiding plane coincides with the marble-schist-interface.In other places, dissolved fissures in the ceiling (severalmetres high) demonstrate that the guiding plane occurswell within the marble layer (Fig. 4).

The passage floor contours make a remarkable bendindicating a change in the dip direction of the foliation


Fig. 6. Present and paleowater flow directions in Oksholadeduced from scallops, vadose canyons and qualitative tra-cing experiments.

8/4/2019 Okshola1

9/17


Fig. 7. The guiding planecomprises in various places a

fracture zone, a few dm thick. Marked, presumably Riedelshear planes (black lines)and interpreted movementdirections (red arrows) inaccordance with these shears.Compass for scale.

Fig. 8. Okshola viewed down dip of the guiding plane (oblique to the hinge line).

Fig. 9. Left (a): Stereograpic projection (equal area, lower hemisphere, magnetic north) of poles to foliation and fractures(logged), and the guiding plane (determined in 3D model). Grey great circle and its pole (hinge line) corresponds to the guid-ing plane. Right (b): Rose diagrams (equal area, magnetic north) of trend of surface fractures, guiding fractures and compassbearings from survey. Circle: 10 %.

8/4/2019 Okshola1

10/17

132

and the guiding plane between southern Okshola andnorthern Kristihola (Fig. 3a). Observations of folia-tion and the guiding plane in Kristihola suggest a dip ofabout 25 towards the NE. This contrasts with the SW- toSSW-trending dip direction of Okshola. To explain theseobservations without postulating complex faulting, the

guiding plane, foliation and rock boundaries are thoughtto comprise a saddle surface (illustrated in Fig. 3c). Thisis consistent with the complex fold interference patternsof the rock sequence shown on the geological map (Fig.2).

Steep guiding fracturesThe prominent set of guiding fractures has a steep dipand trends NNE (Fig. 9a), parallel with the Caledonianstrike direction. However, when guiding fractures areplotted separately for each morphological section, thedominating trend is NNE in the upper areas while it isNNW (to NW) in the lower areas (Fig. 10).

The rose diagram of compass bearings recorded duringthe survey (Fig. 9b), and the conspicuous passage trendsillustrated on the cave map (Fig. 10), indicate E-W toENE-WSW as an additional prominent passage trend.However, this fracture set was only detected in CM. Apossible E-W trending (normal) fault was detected in

the central FMRC-zone (dashed line in Fig. 10) and sup-ported by tightening of contours in adjacent passages(Fig. 3a). However, no fault plane was identified and nodisplacement observed due to dissolution and break-down. Therefore, the orientation of the structure wasdetermined in the 3D-model: 097/40.

Fractures mapped in marble and schist outcrops at theterrain surface (surface fractures) display one prominentset trending N-S and two minor sets trending E-W (toESE-WSW) and NW-SE (Fig. 9b). Surface fractures andguiding fractures have partly overlapping orientations,although the dominant fracture set is slightly displaced.


Fig. 10. Okshola. Great cir-cles of guiding fractures fromeach morphological sectionand orientation of the guid-ing plane from the 3D model(grey). Dashed line: postulated fault plane (097/40) observedin the cave, supported by close-

ness of contours in Fig. 3 anddetermined in the 3D model.

8/4/2019 Okshola1

11/17

133

Discussion

Structural speleology

The folding of the rock sequence is likely the cause of someof the variations in trend of guiding fractures in the various

sections of the cave. A possible explanation for the lack ofrecorded E-W trending guiding fractures (seen on the sur-face) may be that these fracture planes are of limited extentso that, in a mature cave system like Okshola, most of themhave been dissolved away.

In accordance with the large canyons and phreatic conduitsin the upper passage of the FMRC-zone, vadose wideningand draining of phreatic passages are likely causes of theobserved breakdown in the central FMRC-zone. An abun-dance of speleothems (stalactites and stalagmites) in someof the caverns demonstrates that meteoric seepage dissolvesalong fractures in the roof, thus reducing its strength.

Origin of fracturesThe identified and postulated trends of guiding fractures,and thus the two prominent passage trends in Okshola,NNE-SSW and E-W, coincide with the local sets of tec-tonic lineaments in the Fauske area (Gabrielsen et al. 1981;Gabrielsen et al. 2002). Accordingly, the fractures may beof regional tectonic origin, and therefore quite old (Gabri-elsen et al. 1981; Gabrielsen et al. 2002). Based on faultsand fracture systems offshore-onshore in the Lofoten-

Vesterlen area, Bergh et al. (2007) suggested a progres-sive clockwise rotation of the regional stress axes from c.E-W to NNW-SSE in the time interval from the Mesozoicto the Palaeogene. The field area is located about 130 kmSE of this area, and may have been influenced by the sameregional stress regimes. Both sets of vertical guiding frac-

tures could develop as a result of these regimes. This impliesthat fracturing probably took place long before speleogen-esis commenced. The cave system is located less than 260m above sea level (a.s.l.) in a glacially overdeepened fjord

valley, so that a pre-glacial (Tertiary/Neogene) origin forthe vadose parts of the cave is unlikely (Lauritzen 1990).

The low-dip guiding plane is, in general, parallel with thetopography. This surface-parallel attitude would probablyexpose the (presumably sheared) guiding plane to open-ing by pressure release due to erosion and glacial unload-ing. Erosional or glacial unloading may also have resultedin secondary mechanical enlargement of steep guiding

fractures. This would make fractures more favourable forwater penetration, and a large initial aperture has been sug-gested to enhance network formation (Palmer 1975, 1991;Howard & Groves 1995). The present data do not allow achronological differentiation of the fractures to be made,and all observations and analyses suggest that fracturingtook place long before speleogenesis commenced. Nounequivocal evidence of neotectonic deformation has beenobserved within the study area; all recent movements (i.e.rockfalls) can be explained by the effect of gravity. The


Fig. 11. Principal sketch of the relationship between hydrological regimes in the karst aquifer and climatic conditions and glacier thickness during a typical glacial/interglacial cycle (represented by the last one). Marine oxygen-isotope record ofDSDP site 607 for the last glaciation from Raymo & Ruddiman (2004). The FIS and MIS terms are adopted from Kleman& Stroeven (1997), and boundaries are modified to fit the hydraulic conditions in the cave. FIS: Fennoscandian ice sheet

corresponding to maximum global cooling and ice build up; MIS: Mountain ice sheets corresponding to intermediate globalcooling and ice build-up (e.g. Younger Dryas to Pre-Boreal ice extent) (Kleman & Stroeven op. cit.). Inset: Principal sketch ofthe glacier surface slopes above Okshola during two different ice sheet extents.

8/4/2019 Okshola1

12/17

134

fracture pattern can therefore, like the bedrock stratigraphy,be regarded as a passive template for speleogenesis.

Speleogenetic setting

Morphological features and sedimentary deposits demon-

strate the strong variability in hydraulic regimes that haveformed the cave. Distinct phases, each with characteristichydrological conditions, can be identified and related tospecific climatic conditions (Fig. 11). Phreatic cross-sec-tions, paragenetic features, scallops in the walls and ceilingand parallel accretion (silt and clay deposits on walls andceiling) demonstrate that the cave has been water-filled,probably during several stages. Water-filled conditions inthe marble apparently occurred under two different set-tings (Lauritzen 1990); either by bedrock control (prior topuncturing of the karst aquifer at a lower level) or by glacialcontrol (i.e. the water level was raised by a glacier occupyingthe Nedrevatn valley).

Glacial stagesParallel accretion (sediments deposited to ceiling as wellas walls and floor) of silt and clay demonstrates that therehave been very slow or stagnant phreatic flow conditionsduring the latest active stages in the cave. The last water-filled episode was necessarily related to subglacial con-ditions, as the entire Okshola cave was situated above themarine limit (110-130 m a.s.l.). Glacial meltwater containedlarge amounts of rock flour (silt and clay) that settle fromsuspension under stagnant conditions. Stagnant phasesof silting-up are generally associated with thick ice cover,

when the slope of the ice surface is gentle (Ford 1977). Thiscorresponds to the existence of a Fennoscandian ice sheetwith the ice front located over the continental shelf, similarto the last glacial maximum (FIS events, Kleman & Stro-even 1997) (Fig. 11). Parallel accretion, with silt and claystill preserved along the entire conduit perimeter, impliesthat the stage of silting-up lasted quite a long time and thatthe sediments were not removed by later flushing. Thesepassages and sections have therefore not been exposed tosubsequent high flow rates during draining, indicating thatit either did not occur or that these passages were accompa-nied by mature bypass-passages of large draining capacity.

The presence of overturned calcite speleothems and sub-rounded boulders demonstrate that, at least parts of thecave have experienced hydraulic regimes of high energycharacterized by flushing after speleothem precipitation.Episodes of flushing are generally associated with thin, wet-based glacial cover (i.e. water present at the ice base) with asteep surface slope (Ford 1977). This corresponds to phasesof glacial advance and recession, similar to those that tookplace during the Younger Dryas and the Pre-Boreal (MISevents, Kleman & Stroeven 1997) (Fig. 11). During thesestages, the Nedrevatn valley was an efficient drainage chan-nel for the inland ice sheet and the ice front was at the coast

or within the fjord (Andersen 1975). In a thin, wet-basedglacier, the glacial hydrological regime, and thus the subgla-cial karst water circulation, is dictated by seasonal and diur-

nal fluctuations in glacial meltwater (e.g. Bennett & Glasser1996). Flushing probably represented peak floods duringsummer when the subglacial drainage system was fullydeveloped and melting most intense (e.g. Brown 2002).

Sand is the most abundant, water-derived deposit in

Okshola. According to the Hjulstrm diagram (Sundborg1956), sand is deposited when the velocity drops below 20to 40 cm/s. This suggests to us that the sand was depos-ited under the same flow regime during which the scallopsdeveloped (approximate velocity estimate 15-30 cm/s). Thisflow regime represents an intermediate state between stag-nant conditions and flushing. Scallop-dominant dischargecorresponds to the upper 2-15 % of the annual flow regime(Lauritzen 1989b; Lauritzen & Lundberg 2000). Conse-quently, the origin of phreatic scallops and sand deposits isalso thought to relate to the MIS glacial events (Fig. 11). Thelow velocity of phreatic flow revealed by the scallops dem-onstrates that flushing had too short a duration to corrode

new scallops in the walls, i.e. it had no speleogenetic effect.This supports our assumption that flushing represents shortpeak flood events while scallops and sand deposits relate tothe annual flow regime under a thin, wet-based glacier.

A massive flowstone was precipitated over a gravel depositwith rounded cobbles on a ledge in upper FM (the FM2-sequence; Fig. 12). The flowstone was examined and U/Th-dated by Lauritzen (1995), who found that it grew approxi-mately between 145 and 81 ky BP (i.e. during the Eemianinterglacial). This means that the gravel below it was prob-ably deposited during the previous, Saalian, glaciation. A

silty hiatus within the flowstone represents a cessation ofspeleothem growth, and the top surface represents anotherhiatus due to bulk re-solution of the sample. The silty hiatuswithin the flowstone is interpreted as an episode of stagnantflooding during Termination II at the end of the Saalianglaciation, while the re-dissolved top surface shows the cor-rosional influence of the last glaciation (Weichsel) (Laurit-zen 1995). The occurrence of this sequence on the slopingwall of the phreatic cross-section indicates that this part ofthe FM gallery was formed prior to late Saalian (i.e. graveldeposition) and that the upper FM gallery has not experi-enced phreatic water circulation of much speleogenetic sig-nificance during the Weichselian. The phreatic tube existedprior to the deposit of Saalian gravel.

Breakdown material covered by fine, water-lain sediments(sand and silt) demonstrates that these specific break-down events occurred prior to the last flooding. In conse-quence, they cannot be uniquely related to draining of thecave during the last deglaciation. Present sediment chokesand paragenetic features demonstrate the occurrence ofrepeated episodes of sediment injection and flushing. Asmaller phreatic cross-sectional area in UM may possiblybe explained by this area frequently being a back-waterarea with less flushing and more silting-up and sediment

preservation, thus inhibiting dissolutional wall retreat overlong periods. Additionally, high flow rates through the FMtrunk passage may have enhanced widening in adjacent and


8/4/2019 Okshola1

13/17

135

downstream conduits. This is supported by scallop mor-phometry which displays an overall flow direction awayfrom the FM passage.

Interglacial stagesThe vadose inlet passages display topographically hang-ing positions. They have negligible, if any, drainage areasdespite the large dimensions of their canyon sections. Thedendritic plan and size variation of the vadose passages,together with U/Th-dated speleothems (>350 ka, Laurit-zen 1996), indicate that the cave system has a long, multi-stage history of evolution under open-channel conditionsrelated to interglacials and interstadials (Lauritzen 1991b)(Fig. 11).

The size of the lower vadose canyons (FMRC and EI)implies that they have carried considerably higher flow

volumes over longer time spans than the upper vadose pas-sages (NR and UE). The Kristihola Streamway was sub-merged below sea level during the last deglaciation (cf. ML:110-130 m above present sea level, Fig. 1b). This indicatesthat the evolution of the Streamway was not related to anice-marginal situation during deglaciation, but to a fullinterglacial climate when the sea level was lower than thelowest parts of the canyon system of Kristihola (30 m a.s.l.).

Moreover, flooding events in the Okshola River, which arecapable of filling the whole width of the largest vadose can-yons, demonstrate that the present drainage areas and flow

regimes are large enough to have formed both the Stream-way and the Vestibule passage.

Concerning the inlet passages of NR and UE, the presentrunoff from the hillside is disproportionate and far toosmall to have incised these canyons. If the entire tributary

valley drained into the NR passage the drainage area wouldbe about 90 % of the present, and the corresponding flowrates would be about 90 % of the present Okshola River.The canyons of both NR and UE seem too small to havedrained so much water. We therefore suggest that these can-yons either formed by conveying only part of the flow ofthe Okshola River, or they represent stages when the catch-ment was smaller. Simultaneous development of the paral-lel vadose streamways is doubtful because available run-offfrom the drainage area seems to be too small, and the differ-ence in altitude too large, being 80 m over 400 m betweenNR and the river course at the top of the stream sink doline.Therefore, sequential evolution of the vadose inlet passagesseems more likely.

If the vadose incision in the Kristihola Streamway (10-30 mdeep) developed during the present and the previous inter-glacial (about 75 ka in accordance with speleothem growth,Lauritzen 1995), this would require a rate of down-cutting

in the range of 0.1 to 0.4 m/ka (= mm/a). This is within therange of corrosion rates estimated in present cave streamselsewhere in Nordland: 0.2-0.6 mm/a in Glomdal (Lau-


Fig. 12. The FM2-sequence. An Eemian flowstone overlies gravel with subrounded pebbles and boulder of presumed Saalianage. Hand and ruler for scale. Photo: S.E. Lauritzen.

8/4/2019 Okshola1

14/17

136

ritzen unpub.). Accordingly, it is possible that the incisionof the Streamway, and thus the stream sink doline and EI,developed mainly during the Eemian and Holocene inter-glacials.

Speleogenesis in association with glacial landscape

evolutionThe morphological features and sedimentary deposits inOkshola demonstrate that there have been cyclic changesin the hydraulic regime. This emphasizes a complex multi-stage development through several glacial/interglacialcycles. Consequently, we suggest that the speleologicalevolution occurred together with the glacial erosion of thelandscape (Fig. 13).

In stripe karst, and other types of contact karst, glacial ero-sion and valley entrenchment are likely to cause a process ofcaprock (or wallrock) stripping. When glacial erosion low-

ers the valley floor, the existing stream sink will be aban-doned in a hanging position in the hillside and a new streamsink will form in the newly exposed carbonate rock. Theorientation and attitude of the carbonate outcrop (lithologiccontact) determine whether the sinks migrate upstream ordownstream. As a consequence, the complex geometry ofthe carbonate layer in combination with (glacial) erosion

has permitted allogenic stream sinks to migrate down-stream in contrast to a purely fluvial system over a homo-geneous substrate where migration of sinks would be in theupstream direction due to fluvial headwater erosion (Ford& Williams 2007). Glacial valley incision in the fjord valley(Nedrevatn valley) is likely to have breached the aquifer at

progressively lower elevations and thus allowed springs toemerge at successively lower positions.

Glacial valley incision (and erosion) is likely to have inten-sified post-MPT (mid Pleistocene Transition) when glacia-tions of larger extent and longer duration occurred (FIS gla-cial events; Fig. 13). Quantitative valley incision-rate stud-ies in alpine settings indicate that most of the present reliefwas formed under the regime of 100 kyr cycles (Lauritzen &Gascoyne 1980; Haeuselmann et al. 2007). We suggest thatphreatic initiation and early development of the labyrinthalstructure occurred prior to breaching of the karst aquifer,i.e. opening of (vadose) flow paths towards a lower base lev-

els in the Nedrevatn valley. Pre-MPT, before the aquifer wasbreached in the Nedrevatn valley, both the interglacial riverand subglacial water flow may have fed the exposed mar-ble and guiding plane in the bottom of the tributary val-ley, whilst phreatic water flow along the hillside may haveinitiated the labyrinthal structure. During glaciations post-MPT, the cave system was probably exposed to stronger


Fig. 13. Synthesis showing main stages of development of Okshola cave. Top: Okshola cave. Active cave conduits (black) with

arrows showing water flow direction. Middle: Suggested evolution of base level during the last 1.5 Myr. Bottom: Oxygenisotope curve from Raymo & Ruddiman (2004), FIS and MIS boundaries from Kleman & Stroeven (1997). MPT = Mid-Pleistocene Transition (Ruddiman et al. 1989).

8/4/2019 Okshola1

15/17


fluctuations in the subglacial flow regime associated withshifts in hydrological settings between dissolutional wallretreat, sediment injection, silting-up and flushing.

Suggested sequence of speleogenetic events

We propose the following simplified evolutionary history

of Okshola to account for the observed morphological andsedimentological features within the cave (Fig. 14).

Stage 1 Glacial and interglacial pre-MPT: Phreatic ini-tiation and early development of the labyrinthalstructure along the guiding plane, close to the hill-side. Pyrite oxidation and unloading by erosionwere possibly important factors initiating the spe-leogenesis. Glacial erosion stripped off the caprockand eventually breached the karst aquifer at a lower

position (Figs. 14a and e).Stage 2 Interglacial NR stage: The allogenic river invaded

the cave system (having a smaller volume than the

Fig. 14. Left column, A-D) Suggested scenarios in the evolution of the Okshola cave, based on section from NS-profile (Fig.

1b). Inset A) shear plane. Blue: marble. Green: schist. Vertical axis exaggerated. Right-hand column, E-G) DEM modelshowing the corresponding stepwise exposure of marble outcrops (green). Numbers denote stream sinks and correspondingsprings (with letter) (1-oldest, 4-active at present). Contour interval: 100 m.

8/4/2019 Okshola1

16/17

138 R. . Skoglund & S.-E. Lauritzen NORWEGIAN JOURNAL OF GEOLOGY

present Okshola River) and established an under-ground flow path presumably towards the SW(probably the shortest distance) (Figs. 14b and e).

Stage 3 Glacial: Phreatic conditions in the aquifer with dis-solutional widening and highly variable hydraulicregime in concert with shifts in glacier thickness

and extent. Glacial incision lowered the floor of thetributary valley and left the previous vadose inlet ina hanging position in the hillside. Erosion exposedfresh marble in the Nedrevatn (fjord) valley andthus punctured the aquifer at progressively lowerpositions.

Stage 4 Interglacial UE stage: The allogenic river invadedthe UE passage (with a lower discharge than thepresent river).

Stage 5 Glacial: Similar to stage 3.Stage 6 Interglacial FM stage: The allogenic river invaded

the FM passage. High flow rates and high capacitythroughout the system correspond to a shift in

the spring position towards the SE, i.e. opening ofthe Streamway passage. Development of IM andenlargement of several parallel trunk passagesmay relate to breakdown events and a voluminoussupply of sediments from breakdown and glacialinjection (Figs. 14c and f).

Stage 7 Glacial: Similar to stage 3.Stage 8 Interglacial EI stage: A new stream sink was

established in the exposed marble in the valleyfloor and down-cutting occurred in the Icehalland in the Streamway. Flow rates were high, witha steep and steady flow towards sea level. Vadose

undermining, collapse and dissolution of break-down material resulted in doline development atthe stream sink (Figs. 14d and g). Kristiholas pres-ent entrance was formed by collapse into the mainstreamway.

Conclusions

Okshola and the adjacent Kristihola display clear signs ofa complex, multi-stage development that probably tookplace throughout most of the Pleistocene. We believe thatthe simplest speleogenetic model to meet all morpho-logical observations is a process of progressive erosionalstripping of overlying caprock mica schist in concertwith glacial erosion. Consequently, the Okshola laby-rinth is older than the main passages of Kristihola, whichdeveloped only after valley erosion had intersected thecarbonates at successively lower elevations of which thepresent karst spring just above sea level is the last stage.

Our present knowledge does not permit a chronologi-cal differentiation of individual guiding fractures, apartfrom the fact that in a metamorphic (marble) setting, allkarst voids are younger than the guiding fractures which

then function as static templates. We have, however,demonstrated that most of the labyrinth is guided by thefoliation parting which might have been opened by shear

and/or unloading. We have no unequivocal observationsof faulting per se in the cave system. Although some frac-ture zones display dense internal fracturing, we have notbeen able to detect mesoscopic sense, nor magnitude ofmovement along any of them.

The geometry of the cave system is dictated by threestructural elements: the (presumably thrusted) foliation-parallel fracture zone, and at least two sets of steeplydipping fractures. Upper Okshola and parts of its lowerentrance area display relatively simple morphology in theform of intact phreatic tubes. The central Fata Morgana-River Canyon zone (FMRC) differs by having distinctcollapse and fractured zones.

In order to explain the observed twist of this foliation-parallel guiding plane between Okshola proper andKristihola, one has to postulate an interference pattern ofrather tight folding (Fig. 3).

This model explains all our observed variations in frac-ture and cave geometry between Okshola and Kristiholaand within zones in Okshola itself.

Acknowledgements The project was funded by The Research Councilof Norway (NFR), grant no. 160232/V30 Porosity development inmarble stripe karst. We wish to thank everyone taking part in thesurvey as field assistants: H. Skoglund, T. Solbakk, S. vrevik, G. vre-

vik, R. Solbakk, N. Ringset, H.. Aarstad, T.I. Korneliussen and K.Mjelle. We also wish to thank U. Holbye for interesting discussions, T.Solbakk for commenting on an early version of the manuscript and thelate M. Talbot for correcting the English of the manuscript. T. Atkinson,A. Palmer and W. White are thanked for their constructive commentsthat improved the manuscript.

References

Andersen, B.G. 1975: Glacial geology of northern Nordland, NorthNorway. Norges Geologiske Underskelse Bulletin 320, 1-74.

Bennett, M.R. & Glasser, N.F. 1996: Glacial Geology: ice sheets andlandforms. John Wiley & Sons Ltd, Chichester. 364 pp.

Bergh, S.G., Eig, K., Klovjan, O.S., Henningsen, T., Olesen, O. & Han-sen, J.A. 2007: The Lofoten-Vesterlen continental margin: a mul-tiphase Mesozoic-Palaeogene rifted shelf as shown by offshore-onshore brittle fault-fracture analysis. Norwegian Journal of Geology

87, 29-58.Brown, G.H. 2002: Glacier meltwater hydrochemistry. Applied Geo-

chemistry 17, 855-883.Curl, R.L. 1986: Fractal dimensions and geometries of caves. Mathe-

matical Geology 18, 765-783.Day, A. 2002: Cave Surveying. British Cave Research Association, Bux-

ton. 40 pp.Feder, J. 1988: Fractals. Plenum Press, New York. 283 pp.Ford, D.C. 1965: A method of contouring cave maps. The National

Speleological Society Bulletin 27(2), 55-58.Ford, D.C. 1977: Karst and glaciation in Canada. In Proceedings, 7th

International Speleological Congress, Sheffield. 188-189.Ford, D.C. & Williams, P. 2007: Karst hydrogeology and geomorphology,

2nd edition. John Wiley & Sons Ltd, Chichester. 562 pp.

Gabrielsen, R.H., Braathen, A., Dehls, J. & Roberts, D. 2002: Tectoniclineaments of Norway. Norwegian Journal of Geology 82 (3), 153-174.

Gabrielsen, R.H., Ramberg, I.B., Mrk, M.B.E. & Tveiten, B. 1981:

8/4/2019 Okshola1

17/17


Regional geological, tectonic and geophysical features of Nordland,Norway. Earth Evolution Sciences 1, 14-26.

Gabrielsen, R.H., Faleide, J.I., Pascal, C., Braathen, A., Nystuen,J.P., Etzelmuller, B. & ODonnell, S. 2010: Latest Caledonian toPresent tectonomorphological development of southern Norway.

Marine and Petroleum Geology 27, 709-723, doi: 10.1016/j.marpet-geo.2009.06.004.

Glover, R.R. 1972: Optical brighteners - a new water tracing agent.Transaction of Cave Research Group of Great Britain 14 (2), 84-88.

Gustavson, M., Cooper, M.A., Kollung, S. & Tragheim, D.G. 2004:Berggrunnskart Fauske 2129 IV, 1:50 000. Norges GeologiskeUnderskelse.

Haeuselmann, P., Granger, D.E., Jeannin, P.Y. & Lauritzen, S.E. 2007:Abrupt glacial valley incision at 0.8 Ma dated from cave deposits inSwitzerland. Geology 35 (2), 143-146.

Harland, W.B. 1956: Exfoliation joints and ice action.Journal of Glaci-ology 18, 8-10.

Heap, D. 1969: Report of the British Speleological Expedition to ArcticNorway 1969, including the work of 1968 Hulme Schools Expedi-tion. 37 pp.

Horn, G. 1937: ber einige Karsthhlen in Norwegen. Mitteilungenfr Hhlen und Karstforschung, 1-15.

Howard, A.D. & Groves, C.G. 1995: Early development of karst sys-tems: 2. Turbulent flow. Water Resources Research 31 (1), 19-26.

Japsen, P. & Chalmers, J.A. 2000: Neogene uplift and tectonics aroundthe North Atlantic: overview. Global and Planetary Change 24(3-4), 165-173.

Kleman, J. & Stroeven, A.P. 1997: Preglacial surface remnants andQuaternary glacial regimes in northwestern Sweden. Geomorpho-logy 19, 35-54.

Kusumayudha, S.B., Zen, M.T., Notosiswoyo, S. & Gautama, R.S. 2000:Fractal analysis of the Oyo River, cave systems, and topography ofthe Gunungsewu karst area, central Java, Indonesia. Hydrogeology

Journal 8 (3), 271-278.Lauritzen, S.E. 1982: The paleocurrents and morphology of Pikhg-

grottene, Svartisen, North Norway. Norsk Geografisk Tidsskrift 36,

183-209.Lauritzen, S.E. 1986: Kvithola at Fauske, northern Norway - an exam-

ple of ice-contact speleogenesis. Norsk Geologisk Tidsskrift 66 (2),153-161.

Lauritzen, S.E. 1989a: Shear, tension or both- A critical view on theprediction potential for caves. In: Proceedings, 10th InternationalSpeleological Congress, Budapest. 118-120.

Lauritzen, S.E. 1989b: Scallop dominant discharge. In: Proceedings,10th International Speleological Congress, Budapest. 123-124.

Lauritzen, S.E. 1990: Tertiary caves in Norway: a matter of relief andsize. Cave Science 17(1), 31-37.

Lauritzen, S.E. 1991a: Karst resources and their conservation inNorway. Norsk Geografisk Tidsskrift 45, 119-142.

Lauritzen, S.E. 1991b: Uranium series dating of speleothems: A glacial

chronology for Nordland, Norway, for the last 600 ka. Striae 34,127-133.

Lauritzen, S.E. 1995: High-resolution paleotemperature proxy recordfor the last interglaciation based on Norwegian speleothems.Quaternary Research 43 (2), 133-146.

Lauritzen, S.E. 1996: Karst landforms and caves of Nordland, NorthNorway. Guide for excursion 2; Climate Change; The Karst Record.160 pp.

Lauritzen, S.E. 2001: Marble stripe karst of the Scandinavian Caledo-nides: An end-member in the contact karst spectrum.Acta Carso-logica 30 (2), 47-79.

Lauritzen, S.E. 2002: Kompendium i grottekartlegging. Norsk Grotte-blad 39, 3-36.

Lauritzen, S.E. 2004: Grottolf Program for processing, plotting and

analysis of cave survey data.Lauritzen, S.E. & Gascoyne, M. 1980: The first radiometric dating ofNorwegian stalagmites; evidence of pre-Weichselian karst caves.

Norsk Geografisk Tidsskrift 34 (2), 77-82.Lauritzen, S.E. & Lundberg, J. 2000: Solutional and erosional morph-

ology of caves. In: Klimchouk, A.B., Ford, D.C., Palmer, A.N. &Dreybrodt, W. (eds.): Speleogenesis: Evolution of Karst Aquifers,406-426. National Speleological Society, Huntsville, Alabama.

Ottesen, D., Rise, L., Knies, J., Olsen, L. & Henriksen, S. 2005: TheVestfjorden-Trnadjupet palaeo-ice stream drainage system, mid-Norwegian continental shelf.Marine Geology 218, 175-189.

Palmer, A.N. 1975: The origin of maze caves. The National Speleologi-cal Society Bulletin 37(3), 56-76.

Palmer, A.N., 1991, Origin and morphology of limestone caves.Geological Society of America Bulletin 103, 1-21.

Pasini, G., 2009: A terminological matter: paragenesis, antigravitativeerosion or antigravitational erosion? International Journal of Spe-leology, 38 (2), 129-138.

Raymo, M.E. & Ruddiman, W.F. 2004: DSDP Site 607 Isotope Data andAge Models. IGBP PAGES/ World Data Center for Paleoclimatology,ftp://ftp.ncdc.noaa.gov/pub/data/ paleo/contributions_by_author/raymo1992/raymo1992.txt [accessed 19.02.09]

Renault, P. 1968: Contribution a ltude des actions mchaniques etsdimentologiques dans la Splogense.Annales de Splologie 23,529-593.

Ruddiman, W.F., Raymo, M.E., Martinson, D.G., Clement, B.M. &Backman, J. 1989: Pleistocene Evolution: Northern Hemisphere IceSheets and North Atlantic Ocean. Paleoceanography 4, 353-412.

Stephens, M.B., Gustavson, M., Ramberg, I.B. & Zachrisson, E. 1985:The Caledonides of central-north Scandinavia - a tectonostrati-graphic overview. In: D.G. Gee & Sturt, B.A. (eds.): The Caledo-nide Orogen - Scandinavia and Related Areas, Part 1, 135-162. JohnWiley & Sons Ltd.

Sundborg, A. 1956: The river Klarlven, a study of fluvial processes.Geografiska Annaler 38, p. 197.

Talwani, M. & Eldholm, O. 1977: Evolution of the Norwegian-Green-land Sea. Geological Society of America Bulletin 88 (7), 969-999.

Okshola1

Documents

Transcript of Okshola1