Uranium-series dating of speleothems from Demanova Ice Cave: A step to age estimation of the Demanova Cave System (the Nizke Tatry Mts., Slovakia)

Annales Societatis G eologorum Poloniae (1997), vol. 67: 439-450.

URANIUM-SERIES DATING OF SPELEOTHEMS FROM DEMÄNOVA ICE CAVE: A STEP TO AGE ESTIMATION

OF THE DEMÄNOVA CAVE SYSTEM (THE NIZKE TATRY MTS., SLOVAKIA)

Helena HERCMAN1, Pavel BELLA2, Jerzy GŁAZEK3, Michał GRADZIŃSKI4, Stein-Erik LAURITZEN & Reidar L0VLIE6

1 I n s titu te G e o lo g ic a l S c ie n c e s, P o lish A c a d e m y o f S c ie n c e s, T w a rd a 51/55, 0 0 -8 1 8 W a rsza w a , P o la n d A d m in is tr a tio n o f S lo v a k C aves, H o d z o v a I I , 031 01 L ip to v s k y M iku ló ś, S lo v a k ia

I n s titu te o f G e o lo g y , A d a m M ic k ie w ic z U n iversity, M a k ó w P o ln y c h 16, 6 1 -6 0 6 P o zn a ń , P o la n d 4 I n s titu te o f G e o lo g ic a l S c ie n c e s, J a g ie llo n ia n U n iversity, O le a n d ry 2a, 3 0 -0 6 3 K ra k ó w , P o la n d

5 In s titu te o f G eo lo g y, B e r g e n U n iversity, A lle g a tte n 41, 5 0 0 7 B e rg e n , N o r w a y In s titu te o f S o li d E a rth P h ysics, B e r g e n U n iversity, A lle g a tte n 41, 5 0 0 7 B e rg e n , N o r w a y Hercman, H., Bella, P., Głazek, J., Gradziński, M., Lauritzen, S.-E. & L0vlie, R., Uranium-series dating of speleothems from Demänova Ice Cave: A step to age estimation o f the Demänova Cave System (the Nizke Tatry Mts., Slovakia). Ann. Soc. Geol. Polon., 67: 439—450.

A bstract: The Th/U and U/U datings indicate 4 episodes of speleothem growth in Demänova Ice Cave, namely:

ca. 685-410 ka, ca. 170-140 ka, ca. 104-70 ka, and < 5.6 ka. The speleothems studied are confined to the IV cave level of the Demänova Cave System. Taking into account the commonly accepted rules o f cave level formation, one should accept that the level IV must have been dewatered before its oldest speleothems developed, i.e. before ca. 685 ka. Since these speleothems are underlain by fluvial sands o f normal magnetic polarity, it is possible to constrain the age o f level IV as falling into the time-span o f 780-685 ka. It means that this level is older than hitherto supposed and, consequently, that the age of higher levels (V-IX) is older as well.

A bstrakt: Datowania Th/U i U/U dowodzą, że w Demenovskiej Lodowej Jaskini czterokrotnie dochodziło do wzrostu nacieków. Stwierdzone zostały generacje o następującym wieku: ok. 685-410 ka, ok. 170-140 ka, ok.

104-70 ka i młodszym od 5.6 ka. Badane nacieki występowały na IV poziomie jaskiniowym w Demenovskim Systemie Jaskiniowym. Biorąc pod uwagę ogólnie przyjmowane zasady tworzenia się poziomów jaskiniowych należy uznać, że IV poziom został odwodniony przed powstaniem najstarszych nacieków na nim występujących.

Stwierdzony U/U wiek tych nacieków wynosi ok. 685 ka. Biorąc pod uwagę, że bezpośrednio pod tymi naciekami znajdują się rzeczne piaski o normalnym namagnesowaniu można określić wiek odwodnienia IV poziomu jaskiniowego na między 780 ka a ok. 685 ka. Dowodzi to, że wiek tego poziomu jest starszy niż dotychczas sądzono. Starszy też musi być wiek wyższych poziomów (V-IX).

Key words: U-series dating, speleothems, cave evolution, Slovakia, Demänova Cave System.

M anuscript received 20 N ovember 1997, accepted 16 Decem ber 1997

INTRODUCTION

T he D em än o v a C ave S ystem (D C S; in Slovak “D em ä- novsky ja sk y n n y system ”) has been described in a m ono­

graph by D roppa (1957) and in num erous subsequent papers by this author (D roppa, 1963, 1966, 1972) as an exam ple o f a m ulti-level cave system , w hose individual levels could be correlated w ith fluvial terraces (see Fig. 2). This system is a m odel exam ple o f caves o f such a type, quoted in textbooks all over th e w orld {cf. S w eeting, 1973, pp. 151-154, fig. 72;

W arw ick, 1976, pp. 112; Bögli, 1980, pp. 118-119, fig. 8.2;

Jennings, 1985, pp. 2 4 2 -2 4 4 , fig. 89). M ore recent, detailed

studies indicate, how ever, th at th e origin o f this system is a m ore com plicated one (H ochm uth, 1988, 1993, 1995; B ella, 1993, 1996). T he dating o f age b oundaries o f this system by independent physical m ethods is o f crucial im portance for further discussion. The m ost ap propriate m ethods in this re ­ spect are palaeom agnetic studies and isotopic datings o f speleothem s.

T he results o f num erous studies indicate th at speleo­

them s can successfully be dated q u antitatively by isotopic m ethods (cf. W hite, 1988; F ord & W illiam s, 1989; Ivano-

440

Fig. 1. Location o f the Demänova Cave System in the Nizke Tatry Mts., arrow indicates Demänova Ice Cave

vich & H arm on, 1992). Such determ inations m ake it po ssi­

ble to recostruct im portant environm ental changes like, for instance, clim atic w arm ings (T hom pson et a l., 1974; H ar­

m on e t a l., 1975; A tkinson e t a l ., 1978; G łazek & H arm on,

1981; G ascoyne e t al., 1983; H enning e t al., 1983; G ordon et a l., 1989; H ercm an, 1991; B aker e t al., 1993; Lauritzen, 1995). The age o f speleothem s and p alaeom agnetic determ i­

nations o f cave sedim ents are also useful in reconstructing episodes o f geom orphic evolution o f a cave-bearing area (Ford, 1973; Ford e t a l., 1981; Schm idt, 1982; W illiam s et a l., 1986; H ercm an, 1991).

Prelim inary results o f this studies w as p resented on 12th International C ongress o f Speleology (H ercm an e t al., 1997).

GEOLOGIC AND

GEOMORPHOLOGICAL SETTING

The D CS is situated in the N izk e T atry M ts., on the eastern side o f the D em änova V alley (in Slovak

“D em änovska d olina” ; D roppa, 1957; Fig. 1). T he system is developed w ithin A nisian lim estones and dolom ites o f the G utenstein type. T hese lim estones belong to the al- lochthonous K riżna sequence th at constitutes the northern sedim entary cover o f the N izk e T atry M ts. crystalline core, the latter being com posed o f granitoids (cf. D roppa, 1957).

The upper part o f the D em än o v a V alley w as glaciated at least tw ice during the M iddle P leistocene. T he DCS, how ­ ever, occurs in a narrow canyon located dow nstream the glaciated part o f the valley and below the preserved till de­

posits (D roppa, 1972). The total length o f D CS attains 24 km and its re lie f is up to 173 m (B ella, 1993). D roppa (1957) distinguished 9 cave levels in the D CS (Fig. 2). In sub­

sequent papers, this author (D roppa, 1963, 1964, 1966, 1972) correlated these levels w ith fluvial terraces o f the D em änovka stream , as w ell as w ith those o f th e V âh R iver and its tributaries. He assigned individual cave levels to suc­

cessive glacial stages, using th e classical A lpine m orphos- tratigraphic schem e.

The D em änova Ice C ave (called in this paper D IC ; in Slovak “ D em änovska Iadova ja sk y ń a ”) form s the northern (resurgence) part o f D CS and is developed in three levels:

IV, V, and VI (D roppa, 1957, 1972; Fig. 2). T he gross part o f the cave is occupied by level IV, w hich is situated ca. 45 m above the present-day D em änovka stream . It is im portant to note th at the level IV is clearly discernible both in DIC and in other parts o f D C S, and th at it is the m ost distinct from am ong all the cave levels in th at area. U nfortunately,

1000

9 0 0

BOO

SOO 1C

I - VIII c o v * le v t l*

p r*s«n t volley bottom

1500 25 0 0 3 5 0 0 bOOOm

Fig. 2. Longitudinal section o f Demänova Cave System, I—IX cave levels are visible (after Droppa, 1966, simplified); arrow indicates the Zâvrtovÿ dóm chamber - sampling place

U R A N IU M -SE R IE S D A T IN G OF SPELEOTHEM S

441

F ig. 3. C ro ss sectio n o f th e Z â v rto v ÿ dóm , nu m b ers in circles in d icate co lle ctio n p laces o f p a rticu la r flo w sto n e sam ples

the level cannot be correlated w ith any preserved terrace o f the D em änovka stream . D roppa (1966) relates the origin o f this level to th e M indel II glacial stage, basing on a co rrela­

tion w ith terraces preserved in valleys o f the V âh and other tributaries o f V âh River.

MATERIAL

F low stone sam ples have been collected in a cham ber called Z âvrtovÿ dóm o f DIC , belonging to the IV cave level distinguished by D roppa (1957, 1966). G eologic setting o f this cham ber is show n in Fig. 3. Seven flow stones, o ccur­

ring side by side or in superposition have been sam pled.

T hese flow stones are intercalated by clastic sedim ents.

S am ple 1, 2 and 3 w ere collected from fractured, rotated and slightly displaced flow stones, w hich are a very charac­

teristic elem ent o f the cham ber (Fig. 4). Sam ples 4 and 5 w ere tak en from flow stones occurring inbetw een clastic sedim ents in the w estern part o f the cham ber, w hereas sam ­ ples 6 and 7 w ere collected from flow stones w hich drape over a block lying in the m iddle o f the cham ber (Fig. 5). All the sam ples have been labelled in field JL o d l through JL od

F ig . 5. B lock ly in g in th e cen tre o f the Z âv rto v ÿ dó m , nu m b ers in d icate co llectio n p laces o f p a rticu la r flo w sto n e sam p les

7. For sim plicity purposes, in th e rem aining part o f this p a ­ per, w e shall use only sam ple num bers, w ith o u t letter codes.

METHODS

S edim entological analyses included exam ination o f polished slabs and thin sections o f specim ens. T hin sections have been studied u nder an A xioskop C arl Z eiss O pton p ét­

rographie m icroscope, coupled w ith a M C 80 photo cam era.

For radiom etric dating, standard radiom etric d ating o f

i'xa.

T h/ U dates w ere used (cf. Ivanovich & H arm on, 1992). Sam ples o f 15 -3 0 g w ere dissolved in ca. 6 M nitric acid. U ranium and thorium fractions w ere separated by the chrom atography m ethod. The 234U , 238U, 23 Th and 2Th activities w ere m easured by using isotope dilution w ith 228T h/232U spike. A ll m easurem ents w ere done w ith alpha spectrom etry at the U -S eries L aboratory in B ergen U n iv er­

sity. The ages w ere calculated by a standard algorithm (cf.

Ivanovich & H arm on, 1992) using program “A ge04"

(Lauritzen, 1981). T he reported errors are 1 sigm a. A fter d a tin ^ a lH h e sam ples by m eans o f th e 230T h/23 U m ethod, the U / U m ethod w as used to estim ate th e age o f the oldest flow stones from D IC , according to R U B E m ethod (Ivanovich & H arm on, 1992). T he calculated initial - 34U /238U ratios in the younger sam ples w ere used for esti-

234 23o

m ation o f the initial U / U ratio in sam ple 4. W e used 2^4 238 the m ean value o f all the calcu lated initial ' U / ' U ratios

234 238

and obtained estim ator o f initial " U / ' U in th e sam ple 4 equal to 2.62+/-0.38.

P alaeom agnetic analysis included m easurem ents o f natural rem anent m agnetization (N M R ) o f 11 subsam ples (volum e co. 24^10 m m 3), cu t from sam ple 4, on a C ry o ­ genic m agnetom eter (sensitivity 0.02 m A m ’1).

RESULTS OF URANIUM SERIES DATING

All th e obtained results are reliab le due to the high enough uranium content and th e lack o f detrital thorium contam ination (Table 1). T he results o f 230T h/234U dating F ig . 4. G en eral view o f the Z â v rto v ÿ dóm , n u m b ers indicate

c o lle ctio n p lac es o f p articu lar flo w sto n e sam ples

442

Table 1

230T h/23^U dating results o f flow stone sam ples

Sample Cone. U

[ppm]

234u/238u

[ka}

1/1 0.28±0.006 1.408±0.026 0.837±0.041 > 10000

+20 169

-17

1/3 0.38±0.009 1.416±0.032 0.836±0.029 1015

+ 14 169

-12

1/6 0.15±0.004 1.535±0.048 0.791±0.029 > 10000

+11 148

-8

1/9 0.17±0.005 1.460±0.036 0.764±0.028 > 10000

+ 10 140

-9

2/1 2.49±0.08 2.496±0.082 0.548±0.023 >10000

+4.6 78.7 -4.4

2/2 0.36±0.009 1.989±0.046 0.530±0.019 > 10000 +3.8

76.3 -3.7

2/3 0.22±0.005 2.215±0.045 0.499±0.019 > 10000 +3.5

69.9 -3.4

2/4 0.49±0.011 2.896±0.049 0.042±0.003 > 10000 +0.3

4.7 -0.3

3/1 0.16±0.004 1.967±0.539 0.561±0.022 > 10000

+4.6 82.5

-4.4

3/3 0.19±0.004 1.868±0.044 0.518±0.022 > 10000

+4.3 74.2 -4.2

4/1 0.59±0.014 1.181±0.033 1.008±0.038 > 10000 > 350

4/2 0.28±0.010 1.395±0.058 1.043±0.050 436.5 > 3 5 0

5/1 0.30±0.007 2.075±0.041 0.604±0.019 140 +4.2

91.4 -4.0

5/1 a 0.37±0.009 2.134±0.046 0.574±0.019 230.6 +4.2

84.7 -4.0

5/2 0.30±0.006 2.078±0.038 0.548±0.015 662 +3.0

79.5 -2.9

6/1 0.48±0.009 1.975±0.035 0.649±0.016 600 +4.0

102 -3.9

6/2 0.44±0.015 2.100±0.076 0.649±0.030 140.3 +7.4

101 -7.0

7/1 0.42±0.008 2.084±0.043 0.661±0.017 > 10000

+4.2 104

-4.1

7/2 0.29±0.010 2.024±0.072 0.609±0.029 > 10000 +6.7

92.5 -6.4

7/3 0.42±0.009 2.657±0.050 0.051+0.004 > 10000

+0.4 5.6

-0.4

are presented in T ab le 1 (see also Figs. 8 -1 0 ). T able 2 con­

tains results o f 234U /238U dating o f sam ple 4 (see also Fig.

7).

B a sed o f th e datin g results w e can distinguish at least four generations o f speleothem s w hich developed in indi­

v idualised periods. T he first generation (sam ple 4) w as de­

posited betw een ca . 685 k a -4 1 0 ka, th e second (sam ple 1) betw een ca. 1 7 0 -1 4 0 ka, th e th ird (sam ples 2, 3, 5 -7 ) be­

tw een ca. 104-70 ka, and th e fourth one (upper p arts o f sam ­

ples 2, 3, and 7) after 5.6 k a (see Fig. 13).

RESULTS OF PALAEOMAGNETIC ANALYSIS

T he top 5 subsam ples hav e intensities close to the noise level o f th e instrum ent. N o reliable p o la rity has been o b ­ tained from these levels in th e speleothem . T he low er 6 sub-

UR A N IU M -SER IES D A TIN G OF SPELEOTHEM S

443

T a b le 2 R U B E dates o f sam ple 4

S a m p le 2 3 4 U /2 3 8 U age

[ka]

+ 6 0

4/1 6 8 5

- 4 0

+ 73

4 ,2 4 1 0

- 4 6

sam ples have N M R intensities w ell above the noise level, and progressive alternating field dem agnetization to 40 m T has been perform ed. The results indicate the presence o f a single com ponent m agnetization o f norm al polarity (Fig. 6).

-1

I n c l i n a t i o n In te n s ity m A m

-9 0

0

X)2 101 1

0 -

1 -

2 ^-

3 - U -

=* 5 -

•

E 6 -

oM 3 7 - i/>

8 -

9 - 1 0

11 -

Fig. 6. S tratigraphie p lots o f natural rem nant m a g n e tiza tio n in­

c lin a tio n and in tensity; sa m p le 4, f iv e upper su b sa m p les are c a lc ite flo w s to n e , s ix lo w e r are c em en ted c la s tic d e p o sits

F I R S T S P E L E O T H E M G E N E R A T I O N

D e s c r ip tio n . T he first speleothem g eneration is rep re­

sented by a 10 cm -thick flow stone (sam ple 4, Fig. 7), d ev el­

oped upon cave infill w hose top, 8 cm -thick part, is ce­

m ented by calcite. This infill is com p o sed o f siliciclastic m aterial, including grains o f quartz, v ariab ly w eath ered bi- otite, m uscovite and feldspars, as w ell as o f carbonate silt.

The overlying flow stone is com posed in its low er an d u pper parts m ainly o f colum nar and, su bordinately, acicular crys-

FLOWSTONES

■J ■ +73]

- ■ -. t. t - Â ' r 10 ka

..

461

1

s ¹

■ i 1 jF\ ' +OT|

j 685 ka 1 V v ■-r r-

Mm

Fig. 7. The sample 4 represents the oldest speleothem genera­

tion in Demänova Ice Cave, U/U ages (ka ago) o f particular layers are indicated, in the centre o f sample the lens filled with macro- scopic-size calcite crystals that developed in a small pool, is visible

tals. C orrosion surfaces are to be found a t places. T he m id ­ dle part o f the flow stone is occupied by a lense o f porous sedim ent, com posed o f calcite crystals, 3.5 cm high.

In te r p r e ta tio n . T he first sp eleo th em g eneration grew under suitable conditions at unin terru p ted supply o f w ater, as show n by w ell-developed co lu m n ar calcite cry stals (cf.

D ziadzio e t al., 1993; F risia e t a l., 1993; G radziński e t a l., 1996). C orrosion surfaces poin t to tem p o rary interruption o f the grow th and destruction o f p rev io u sly precip itated flo w ­ stones; nevertheless, they do n o t rep rese n t — m o st pro b ab ly - a profound hiatus in th e speleothem grow th. T he porous sedim ent that builds a lense-like bo d y in the m iddle p art o f the flow stone developed in a sm all lake, several centim etres deep (cf. G onzalez & L ohm ann, 1988; G on zalez e t a l., 1992).

S E C O N D S P E L E O T H E M G E N E R A T I O N D e sc rip tio n . T he second speleo th em g eneration is a 12 cm -thick flow stone, yellow in colour (sam p le 1; F ig. 8), d e­

veloped upon cave infill, w hose to p part, 1 cm thick, is c e ­ m ented by calcite. T his flow stone is largely com posed o f colum nar calcite crystals and does n o t show co rrosion sur­

faces. In its m iddle p art a m acro sco p ically distinguishable, lighter layer occurs, show ing subm icro sco p ic lam ination.

A bove a distinct corrosion surface on to p o f th e flow stone, a

444

Fig. 8. The sample 1 represents the second speleothem genera­

tion in Demänova Ice Cave, U/Th age (ka ago) of particular layers are indicated, the layer built of cemented intraclasts is visible in the upper part of the sample (see text for further explanations)

3 cm th ick layer occurs w hich is built up by clasts com posed o f fine-grained sedim ents, ranging from fine-grained sand to silt. T hese clasts are fragm ents o f the cem ented cave infill;

hence, they represent intraclasts. T hey display internal hori­

zontal lam ination. T he clasts are rotated, displaced and ce­

m ented by spary cem ent.

In te r p r e ta tio n . C olum nar calcite crystals b uilding the flow stone represent th e colum nar m icrofacies, distinguished by D ziadzio e t al. (1993) and G radziński e t al. (1996). Their reg u lar d evelopm ent testifies to suitable growth conditions o f th e flow stone in question (see also Frisia e t a l., 1993).

T he lack o f corrosion and nucléation surfaces in the log de­

scrib ed is indicative o f stable hydrodynam ic conditions dur­

ing th e flow stone grow th. L am ination occurring in the m id­

dle p a rt o f the flow stone is associated w ith the cyclic, p ro b ­ ably seasonal delivery o f alternately clear and contam inated by organic and/or m ineral substance solutions (G radziński e t a l., 1997a). A distinct corrosion surface at the top records destructive processes th at had occurred before th e layer com posed o f intraclasts w as deposited. The spatial arrange­

m ent o f intraclasts resem bles th at o f teepe structures (cf.

T ucker & W right, 1990). T he presence o f such structures is indicative o f rapid cem entation o f clastic sedim ents, dis­

p lacem en t o f clasts, m ost certainly at short distances, and re ­ p eated cem entation o f the rew orked sedim ents in stagnant w aters.

Fig. 9. The sample 2 represents the third and the fourth spe­

leothem generations in Demänova Ice Cave, U/Th age (ka ago) of particular layers are indicated, corrosion surface between both generations is visible (big arrows)

flow stones show as w ell differen t internal structure (sam ­ ples 2, 3, 5 -7 ; Figs. 9, 10). A ll o f them w ere grow ing upon cem ented, siliciclastic cave sedim ents. S am ples 2 and 5 rep ­ resent the infill com posed o f ro unded intraclasts, derived from older cave deposits (Fig. 11). The flow stones in ques­

tion are com posed o f acicular and colum nar calcite crystals.

The acicular crystals usually form m acroscopically recog­

nisable, w hite, dom e-like form s. F low stones o f the third generation do also include frequent nucléation surfaces (Fig.

12) w hich are responsible for m acroscopically visible lam i­

nation. Sam ple 3 reveals ano th er character. It show s high porosity and the presence o f acicu lar calcite crystals w hich grow freely at variable angles to th e n u cléation surfaces.

The topm ost parts o f all th e flow stones bear distinct corro­

sion surfaces.

In te rp re ta tio n . The th ird speleothem generation grew

THIRD SPELEOTHEM GENERATION

D e s c r ip tio n . T his generation is represented by flow ­ stones o f variable thickness, from 2.5 to 15 cm. Individual

Fig. 10. The sample 7 represents the third and the fourth spe­

leothem generations in Demänova Ice Cave. U/Tli age (ka ago) of particular layers are indicated, corrosional surlacc between both generations is visible (big arrow)

U R A N IU M -SER IES DA TIN G OF SPELEOTHEM S

445

Fig. 11. Intraclasts comprise clastic materials which build low­

ermost part o f sample 2; X nicols, scale bar 0.8 mm

Fig. 12. Nucléation surface well visible due to competitive crys­

tal growth pattern (cf. Sunagawa. 1994), upper part of sample 2; X nicols, scale bar 0.8 mm

under considerably less stable conditions, as com pared to the previous one. A cicular calcite crystals, being a com m on com ponent o f this generation, belong to the acicular m icro- facies, distinguished by G radziński e t al. (in press). Such crystals originate under unstable conditions, characterised by the presence o f supersaturated solution (see also G iven &

W ilkinson, 1985; G onzalez e t a l., 1992; Jones & K ahle, 1993). N um erous nucléation surfaces testify to frequent in­

terruptions in the speleothem grow th, probably due to dry­

ing up o f the cave. C om paring th e properties and ages o f in­

dividual flow stones, one can conclude that the grow th con­

ditions gradually deteriorated, from relatively suitable ones at the tim e o f developm ent o f th e oldest flow stone (low er part o f sam ple 7), to the less favourable conditions during the grow th o f the youngest flow stone, rich in nucléation sur­

faces (sam ple 2). The flow stone grow th was coeval w ith the developm ent o f sm all lakes. T he infill o f one o f such lakes is represented by sam ple 3 (cf. G onzalez & L ohm ann, 1988;

G onzalez e t a l., 1992).

FOURTH SPELEOTHEM GENERATION

D e s c r ip tio n . This generation is represented by up to 1 cm -thick layers, built up o f m icritic calcite. These layers overlie corrosion surfaces in sam ples 2 and 3 (Figs. 9, 10).

The generation com prises as w ell a flow stone and a sm all stalagm ite, developed upon a corrosion surface on top o f sam ple 7. B oth o f them are com posed o f regularly devel­

oped colum nar calcite crystals.

In te r p r e ta tio n . T he m icritic flow stones are probably the products o f advanced diagenesis o f m oonm ilk (cf. G radz­

iński e t a l., 1997b). The p roperties o f the flow stone and sta­

lagm ite o f sam ple 7 p o int to stable and favourable growth conditions (cf. D ziadzio e t a l., 1993; F risia e t a l., 1993;

G radziński e t al., 1996).

STAGES OF FLOW STONE GROWTH IN DEMÄNOVA ICE CAVE

It w as noted quite early th a t speleothem deposition, at least in th e u pper m id-latitudes, is discontinuous w ith ages clustering into distinct groups th a t correlate b roadly w ith the

know n w arm and hum id stages o f th e Late P leistocene (T hom pson e t a l., 1974; H arm on e t a l., 1975; A tkinson et a l., 1978; G łazek & H arm on, 1981; G ascoyne e t a l., 1983;

H enning e t a l., 1983; G ordon e t a l., 1989; H ercm an, 1991;

B aker e t a l., 1993). Since the speleothem grow th is co n tro l­

led by clim atic factors, w e can com pare th e estim ated stages o f speleothem grow th w ith independent clim atic records.

W e used a geologic tim e-scale developed by Im brie e t al.

(1984), based on analyses o f isotopic d ata from five deep- se a cores. It is notew orthy th at three o f th em pen etrated the B runhes/M atuyam a boundary. T his tim e-scale covers the last 780,000 years.

The oldest stage o f speleothem d ep o sitio n in D IC can be correlated w ith the 20-11 180 stages. T he second perio d is correlated w ith 180 stage 6, w hereas the third one — w ith 180 stage 5. M ore precisely, it can be correlated w ith su b ­ stages 5 c-5 a . The youngest, fourth stage is coeval w ith 180 stage 1 (Fig. 13).

T he above com parison indicates th a t som e speleothem s grew in D IC at the tim e o f “cold” stages. T he best exam ple is provided by flow stones grow ing during 180 stage 6. Sedi-

A .

■ U / U A g e

« U / T h A g e

0 200 400 600 0OO

B.

0 100 200 300 400 500 600 700 BOO

A g e ( h a )

Fig. 13. Correlation o f uranium series dating results o f spe­

leothems from Demänova Ice Cave and oxygen isotope record: A - speleothems dating results, errors la ; B - the stacked, smoothed oxygen-isotope record as a function o f age in SPECMAP time scale. Isotopic variation expressed in standard deviation units around a zero mean (after Imbrie et a i, 1984)

446

5.6 ka

70 ka

104 ka

140 ka

170 ka

410 ka

685 ka

water

sand, silt, clay

speleothems

gravel

Fig. 14. Scheme o f Zâvrtovÿ dóm development from ca. 685 ka to Recent, see text for further explanations

m entological features o f th e flow stones (see above) prove that they grew in favourable, w et and hum id conditions. It suggests that the N izke T atry M ts. w ere situated outside the periglacial zone at that tim e. O n the other hand, the lack o f speleothem deposition during the “w arm ” substage 5e is probably caused by local conditions, b ecause in the other parts o f D CS we have dated flow stones th a t could be corre­

lated w ith th at stage. U ninterrupted flow stone grow th during ,80 stages 20 through 11 is difficult to understand. This problem requires further w ork.

EVOLUTION OF DEMANOVA ICE CAVE

The analysis o f cave passages and th eir infill m akes it possible to reconstruct the evolution o f th e Z âvrtovÿ dóm cham ber in D IC and, basing on this exam ple, the evolution o f the entire IV cave level in DCS (Fig. 14). A ge determ ina­

tions o f cave sedim ents enable us to constrain the ages o f individual stages o f the cave system developm ent.

The developm ent o f the IV cave level should be view ed as proceeding in tw o stages: (1) enlargem ent o f cave p as­

sages, and (2) th e subsequent filling o f such voids by sedi­

m ents. The top passages, occurring in th e top part o f Z âvr­

tovÿ dóm , can be either (1) relics o f incipient p hreatic corri­

dors th at w ere subsequently dissected and rem odelled under vadose conditions (cf. B retz, 1942), or (2) an effect o f p a r - agenesis (Ford & W illiam s, 1989). T here is no sufficient evidence for cave filling up to the top by clastic sedim ents, w hat favours the first option. O w ing to subseq u en t low ering o f elevation o f the spring position, the Z âvrtovÿ dóm cham ­ ber w as placed in the vadose zone, the c h a m b e r’s floor be­

ing system atically low ered. W hen the present-day position o f the floor w as achieved, th e cave becam e infilled by clas­

tic sedim ents, i.e. gravels and sands, as w ell as by flow stone floors. B asing on sedim entary sequences th at fill the cham ­ ber, one can distinguish three stages o f clastic sedim ent deposition and four episodes o f speleothem crystallisation.

T he m aterial o f clastic sedim ents w as derived from the Ni'zke T atry M ts. crystalline core and, p robably, from re­

w orked older or coeval glacial deposits, and transported into the cave by surface stream s. T herefore, deposition o f clastic sedim ents should be linked w ith th e cave flooding, w hereas flow stone form ation on th e cave floor w as associated w ith periods o f cave drying up.

T he first stage o f cave filling is m arked by gravels de­

posited directly on the rocky floor o f the cham ber. T hese gravels are covered by sands (Fig. 14), w hose top part was deposited during th e B runhes C hron, i.e. n o t earlier than 780 ka (cf. Baksi e t a l., 1992; T auxe e t a l., 1992). F ollow ing the cave dew atering, the oldest, i.e. first speleothem generation w as deposited im m ediately upon sandy sedim ents. T he sp e ­ leothem grow th began ca. 685 ka and lasted until 410 ka.

A fter crystallisation o f th e oldest speleothem g enera­

tion, the Z âvrtovÿ dóm ch am b er becam e flooded by w aters w hich laid dow n sands and silts o verlying th e pre-existing speleothem s. A nother episode o f speleothem crystallisation took place betw een ca. 170 k a and 140 k a (second g enera­

tion; Fig. 14). The speleothem grow th w as term inated by an episode o f corrosion, after w hich the cave w itnessed several

U R A N IU M -SE R IE S D A T IN G OF SPELEOTHEM S

447

episodes o f flooding by w aters bringing in clastic m aterial and rew orking older sedim entary infill.

T he block, p resently lyin g in the centre o f the cham ber, probably fell o f f from the roof, either b efore or during d epo­

sition o f th ese sedim ents. T his is indicated by the presence o f 1 cm -thick layer o f siliciclastic m aterial draping over the block surface and locally preserv ed under younger flow ­ stones. T his m aterial w as d eposited in an aquatic environ­

m ent, after the block fell o f f from the ro o f o f the cham ber.

The presence o f clastic sedim ents upon th e block show s that the cham ber m ust have been flooded at least up to the upper edge o f th e block, i.e. the depth o f the w ater basin (a pond?) in the southern part o f the cham ber m ust have exceeded 2 m (Fig. 14).

T he clastic sedim ents w ere covered by another, third generation o f speleothem s b etw e en ca. 104 ka and 70 ka (Fig. 14). T h eir grow th p roceeded u n d er unstable and gradu­

ally deteriorating conditions, associated w ith the system atic cooling during th e advance o f the last glacial cycle.

C rystallisation o f the th ird speleothem generation w as follow ed by: th e form ation o f a pitch in the southern p art o f the cham ber, the fragm entation o f previously crystallised flow stones, and th e displacem ent o f their dism em bered frag­

m ents (Fig. 14). The last process w as caused by erosion and/or gravity sliding o f underlying, uncem ented clastic sedim ents.

T he fragm ented flow stones underw ent intensive ero­

sion, p roducing distinct corrosion surfaces th at are devel­

oped upon the exposed surfaces o f older speleothem s. The corrosion surfaces, in turn, w ere capped in H olocene tim es by th e youngest, fourth generation o f speleothem s.

AN ATTEMPT AT ESTIMATING THE AGE OF THE DEMÄNOVA CAVE

SYSTEM

H orizontal caves o f sm oothed gradients originate in the epiphreatic zone, i.e. at the level o f karst springs or slightly above, depending on h ydraulic gradient (Sw innerton, 1932).

F ord (1977) and F ord and E w ers (1978) show ed th at h o ri­

zontal caves o f sm oothed gradient, also called ideal w a- tertable caves, tend to develop at the boundary betw een the p hreatic and vadose zones, w ithin highly fractured rocks.

T he d evelopm ent o f such caves proceeds at the tim e o f base level stabilisation. D uring rap id valley deepening, the cave level is d ew atered or cut by vadose canyons. D uring the next perio d o f base level stabilisation another, low er, cave level develops. In this w ay, a m ultiphase cave system is being form ed, w h o se individual levels reflect successive periods o f base level stabilisation (F o rd & W illiam s, 1989). C ave levels develop m ore or less at th e form er base level hori­

zons, indicating the position o f form er valley bottom s o f those rivers th a t drained the cave system . The above view is a com m only accepted one, and the one w hich has recently been confirm ed by com plex studies by P alm er (1987) o f nu ­ m erous N o rth A m erican caves. It should also be noted that younger, vertical shafts o f vadose character could truncate th e horizontal, older fragm ents o f cave system s. The origin o f such shafts is associated w ith invasion w aters (F ord &

E w ers, 1978) w hich could form , for instance, d uring glacier m elting (cf. G łazek e t al., 1977).

T he age o f a p articular cave level is d ifficult to estim ate precisely. T he b est approach is to date sedim ents th a t infill th e level, w hose age is the y o ungest possible age o f the level. The age o f speleothem s developed in a given cave level, in turn, gives th e m inim um age o f d ew atering o f such a level. This technique has been applied to the R o ck y M ts.

caves (Ford, 1973; F ord e t al., 1981), th e M am m oth Cave system (Schm idt, 1982), caves in W yan d o tte R idge, Indiana (Pease e t al., 1994), Y orkishire D ale and M endip H ills caves in th e British Isles (A tkinson e t a l., 1978), W estern T atra M ts. caves (H ercm an, 1991), caves n ea r G uilin (W illiam s et al., 1986) and caves o f the B u ch an an K arst in A ustralia (W ebb e t al., 1992). The age o f speleothem s is usually de­

term ined by th e use o f uranium series techniques, w hereas th at o f clastic sedim ents is determ ined by the palaeom ag- netic m ethod.

The D CS is a typical m ultiphase cave system . B asing on concepts o f F ord (1977) and Ford and E w ers (1978), B ella (1993) assigned individual D CS cave levels, distinguished by D roppa (1957), to the ideal w atertable caves or to the m ixture o f phreatic and w atertable levelled caves. T h ere­

fore, it is ju stified to conclude th at th e age o f the oldest spe­

leothem s occurring in the IV cave level o f D IC is th e m in i­

m um age o f dew atering o f this level.

T aking into account that th e age o f th e oldest flow stones o f the IV cave level is ca. 685 ka (see T able 2), one should accept that the level m ust have been already dew atered at that tim e. Fluvial sands underlying the dated flow stone show norm al m agnetic polarity, in d icatin g th at they w ere deposited during the B runhes C hron, i.e. n o t earlier than 780 ka BP (cf. Baksi e t al., 1992; T auxe e t a l., 1992). T h ese data show th a t the dew atering o f the IV cave level occurred b e­

tw een 780 ka and ca. 685 ka. D roppa (1966) associated the origin o f the IV cave level w ith the M indel II glacial stage, basing on correlation w ith fluvial terraces. A com parison o f the estim ated age o f dew atering o f this level w ith th e know n age o f the M indel II glacial stage, dated at ca . 500 ka (K ukla, 1977), calls for revision o f D ro p p a ’s (1966) view.

O ne should also bear in m ind that the discussed age estim a­

tions refer to th e period o f dew aterin g o f th e cave level, hence, the age o f the level itse lf m ust be older.

Episodes o f flooding o f the IV cave level in D C S, leav­

ing siliciclastic sedim ents inbetw een th e 2nd and 3rd and, probably, also betw een the 1st and 2nd speleothem g enera­

tions, should be linked w ith the delivery o f invasion vadose w aters into the cave. T hese w aters originated at th e tim e o f glacier m elting during consecutive glacial stages. It seem s likely, therefore, th at the clastic sedim ents w ere n o t depos­

ited by subterraneous flow s o f th e D em anovka stream , and th at their top cannot m ark the valley b o tto m positio n at that tim e. Sim ilar events o f clastic m aterial deposition w ithin caves, due to huge flow s induced by glacier thaw ing, have already been described from num erous caves like, e.g. those o f the T atra M ts. (G łazek e t a l., 1977), th e M atlock area in D erbyshire (F ord & W orley, 1977) or C astelguard C ave in the C anadian R ocky M ts. (S ch ro ed er & Ford, 1983).

M oreover, the form ation o f a pitch in the southern part o f the cham ber, creeping o f clastic sedim ents and fragm en-

448

tation o f the three older flowstone generations, all resulted from the draining o f invasion vadose waters into lower lev­

els o f the DCS, most probably at the decline o f the last gla­

cial stage.

The above rules governing the development o f multi­

level cave systems and the estimated age o f dewatering o f the IV cave level enable us to infer that before the dewater­

ing, i.e. before ca. 685 ka, the Demanova Valley bottom was situated not higher than 45 m above its present-day position.

Therefore, the rate o f fluvial incision in Late Quaternary times must have been smaller than hitherto supposed. The higher situated DCS levels (V through IX) should, hence, be older as well, and cannot be correlated with successive A l­

pine glaciations, i.e. Mindel I through Donau, respectively, as conjectured by Droppa (1966). We conclude, therefore, that the morphology o f the Nizke Tatry Mts. is older than previously supposed.

The above conclusions are o f preliminary character. A more detailed reconstruction o f the DCS development and its precise dating requires further geomorphological and speleogenetic studies, aided by isotopic datings o f spe­

leothems and palaeomagnetic determinations performed in the remaining cave levels.

CONCLUSIONS

1. Thé IV cave level o f Demänova Cave System became dewatered between 780 ka and ca. 685 ka; hence, it is older than previously supposed.

2. The age o f the remaining cave levels, as well as that o f the Ni'zke Tatry Mts. relief is also older.

3. Demänova Ice Cave bears a record o f four episodes o f speleothem growth, i.e. ca. 685—410 ka, 170-140 ka, 104—70 kä, and after 5.6 ka.

4. Between 410 ka and 5.6 ka, Demänova Ice Cave was flooded at least twice by invasion (proglacial) waters sup­

plied from the melting glaciers.

Acknowledge ments

The research was supported by Polish State Committee for Scientific Research (KBN) grants no. P2/94/0888, 0586/P04/95/09 and 283/P04/96/11. The help in fields o f Slovak and Polish cavers is acknowledged. H.H., J.G. and M.G. are deeply indepted to Slo­

vak Museum o f Nature Protection and Speleology as well as to the Administration o f Slovak Caves for logistic support. Thanks are also addressed to the anonymous reviewers for helpful suggestions that improved the manuscript. M.G. was awarded by the Founda­

tion for Polish Sciences in 1997.

REFERENCES

Atkinson, T. C., Harmon, R. S., Smart, P. L. & Waltham, A. C., 1978. Paleoclimatic and geomorphic implications of 230Th/234U dates on speleothems from Britain. N ature, 272:

24-28.

Baker, A., Smart, P. & Ford, D. C., 1993. Northwest European paleoclimate as indicated by growth frequency variations o f secondary calcite deposits. Paleogeography, Paleoclimato- logy, Paleoecology, 100: 291-301.

Baksjj A.^K., Hsu, V., McWilliams, M. O. & Farrar, E., 1992.

Ar/ Ar dating o f the Brunhes-Matuyama geomagnetic field reversal. Science, 256: 356-357.

Bella, P., 1993. Remarks on the genesis o f the Demänova Cave System. (In Slovak, English abstract). Slovenskÿ Kras, 31:

43-53.

Bella, P., 1996. Geomorphological imortance and problems o f genesis o f Demänovska Cave o f Liberty. (In Slovak, English summary). In: Bella, P. (ed.), Spristupnené jaskyne. Vÿskum, Ochrana, Vyuiivanie. Sprâva slovenskÿch jaskÿn. Liptovskÿ Mikulâs, pp. 46-52.

Bögli, A., 1980. K arst H ydrology a n d Physical Speleology. Sprin- ger-Verlag, Berlin, 286 pp.

Bretz, J. H., 1942. Vadose and phreatic features o f limestone caverns. J. Geol., 50: 675-811.

Droppa, A., 1957. Die H öhlen D em ânovské Jaskyne. (In Slovak, German summary). Vydavatelstvo Slovenskej Akademie Vied, Bratislava, 289 pp.

Droppa, A., 1963. Paralelizäcia riećnych terâs a horizontâlnych jaskÿft. (In Slovak only). G eologicképrâce, 64: 93-96.

Droppa, A., 1964. Die Erforschung der Vâh-Terrassen im mit­

tleren Teile des Kessels Liptovska Kotlina. (In Slovak, Ger­

man summary). G eografickÿ ćasopis, 16: 313-325.

Droppa, A., 1966. The correlation o f some horizontal caves with river terraces. Studies in Speleology, 1: 186-192.

Droppa, A., 1972. Die geomorphologischen Verhältniosse im Demânovskâ-Tal. Slovenskÿ kras, 10: 9—46.

Dziadzio, P., Różniak, R. & Szulc, J., 1993. Origin o f the Pleisto­

cene calcite flowstones o f two caves (Jaskinia Psia and Jaskinia Naciekowa) in the West Tatra Mts. (In Polish, Eng­

lish summary). Przegl. Geolog., 41: 767-775.

Ford, D. C., 1973. Development o f the Canyons o f the South Nahanni River, N.W.T. Canad. J. Earth Sei., 10: 366-378.

Ford, D. C., 1977. Genetic classification o f solutional cave system.

In: Ford, T. D. (ed.), Proceedings o f the 7th International Speleological Congress, S h effield 1977. British Cave Re­

search Association, Bridgwater, pp. 194—196.

Ford, D. C. & Ewers, R. O., 1978. The development o f limestone cave systems in the dimensions o f lenght and depth. Canad. J.

Earth Sei., 15: 1783-1798.

Ford, D. C., Schwartz, H. P., Drake, J. J., Gascoyne, M., Harmon, R.S. & Latham, A. G., 1981. Estimates o f the age o f the existing relief within the Southern Rocky Mountains o f Can­

ada. A ret. Alp. Res., 13: 1-10.

Ford, D. C. & Williams, P. W., 1989. K arst G eom orphology and Hydrology. Unwin Hyman, Boston, 601 pp.

Ford, T. D. & Worley, N. E., 1977. Phreatic cave and sediments at Matlock, Derbyshire. In: Ford, T.D. (ed.), Proceedings o f the 7th International Speleological Congress, Sheffield 1977.

British Cave Research Association, Bridgwater, pp. 194-196.

Frisia, S., Bini, A. & Quinif, Y., 1993. Morphologic, crystal- lographic and isotopic study o f an ancient flowstone (Grotta di Cunturites, Dolomites) - implications for paleoenviron- mental reconstructions. Speleochronos, 5: 3-15.

Gascoyne, M., Schwartz, H. P. & Ford, D. C., 1983. Uranium-se- ries ages o f speleothem from Northwest England: correlation with Quaternary climate. Phil. Trans. Royal Soc. London, B 301: 143-164.

Given, R. K. & Wilkinson, B. H., 1985. Kinetic control o f mor­

phology, composition, and mineralogy o f abiotic sedimentary carbonates. J. Sediment. Petrol., 55: 109-119.

Głazek, J. & Harmon, R. S., 1981. Radiomnetric dating o f Polish cave speleothems: Current results. In: Beck, B. F. (ed.), Pro­

ceedings o f the 8th International Congress o f Speleology, Bowling Green, Kentucky, pp. 424—427.

UR A N IU M -SE R IE S D A T IN G OF SPELEOTHEM S

449

Głazek, J., Rudnicki, J. & Szynkiewicz, A., 1977. Proglacial caves - aspecial genetic type o f caves. In: Ford, T. D. (ed.), Pro­

ceedings o f the 7th International Speleological Congress, Sheffield 1977. British Cave Research Association, Bridgwa­

ter, pp. 215-217.

Gonzalez, L. A., Carpenter, S. J. & Lohmann, K. C., 1992. Inor­

ganic calcite morphology: roles o f fluid chemistry and fluid flow. J. Sediment. Petrol., 62: 382-399.

Gonzalez, L. A. & Lohmann, K. C., 1988. Controls on mineralogy and composition o f spelean carboantes: Carlsbad Caverns, New Mexico. In: James, N. P. & Choquette, P. W. (eds.), Paleokarst. Springer-Verlag, New York, pp. 81-101.

Gordon, D., Smart, P. L., Ford, D. C., Andrevs, J. N., Atkinson, T.

C., Rowe, P. J. & Christopher, N. S. T., 1989. Dating Late Pleistocene interglacial and interstadial periods in the United Kingdom from speleothems growth frequency. Quatern: Res., 31: 14-26.

Gradziński, M., Rospondek, M. & Szulc, J., 1996. Microfacies types o f calcite speleothem: hydrodynamical and chemical controls of their origin. In: Lauritzen, S.-E. (ed.) Climate Change: The K arst Record. Abstracts, Karst Water Institute Special Publication 2, Bergen, p. 45.

Gradziński, M., Rospondek, M. & Szulc, J., 1997a. Environmental controls o f origin o f the annually varved calcite speleothems.

In: Jeannin, P.-Y. (ed.), Proceedings o f the 12th International C ongress o f Speleology, la C haux-de-Fonds, Switzerland, 10- 17.08.1997. Vol. 1. Swiss Speleological Society, Basel, pp.

81-85.

Gradziński, M., Rospondek, M.., Szulc, J., (in press). Paleoenvi- ronmental controls and microfacies variability o f flowstone cover from Zvonivâ cave, Slovakian Karst. Slovak Geological Magazine.

Gradziński, M., Szulc, J. & Smyk, B., 1997b. Microbial agents o f moonmilk calcification. In: Jeannin, P.-Y. (ed.), Proceedings o f the 12th International C ongress o f Speleology, la Chaux- de-Fonds, Switzerland, 10-17.08.1997. Vol. 1. Swiss Spe­

leological Society, Basel, pp. 275-278.

Harmon, R. S., Thompson, P., Schwarz, H. P. & Ford, D. C., 1975.

Uranium-series dating o f speleothems. Nat. Speleol. Soc.

Bull., 37:21-33.

Henning, G. J., Grün, R. & Brunnacker, K., 1983. Speleothems, travertines, and paleoclimate. Ouatern. Res., 20: 1-29.

Hercman, H., 1991. Reconstruction of geological environment on the Western Tatra Mts. based on isotopic dating o f speleo­

thems. (In Polish, English summary). Zeszyty Naukowe Poli­

techniki Śląskiej, M atematyka-Fizyka, 66, Geochronometria, 8: 1-139.

Hercman, H., Bella, P., Gradziński, M., Głazek, J. & Lauritzen, S.-E., 1997. The antiquity o f the famous Demanowska caves (Slovakia). In: Jeannin, P.-Y. (ed.), Proceedings o f the 12th International Congress o f Speleology, la Chaux-de-Fonds, Switzerland, 10-17.08.1997. Vol. 1. Swiss Speleological So­

ciety, Basel, pp. 85-86.

Hochmuth, Z., 1988. Geomorfologickÿ vÿskum a topografia vod- nej cesty medzi jaskyńami Vyvieranie a j. Slobody v Demä- novskej dolinę. (In Slovak only). Slovenskÿ kras, 26: 7-23.

Hochmuth, Z., 1993. Results o f a detailed mapping and revised geomorphpological survey of the Cave Vyvieranie in the Demänova Valley. (In Slovak, English summary). Slovenskÿ kras, 31: 29-42.

Hochmuth, Z., 1995. Some notes concerning the research o f the phreatic zone in the cave system o f Demänova valley (Low Tatras, Slovakia). In: Bella, P. (ed.), Caves and Man. Interna­

tional sym posium on the occasion o f the 70-th anniversary opening in the public o f the D em änovskä cave o f Liberty,

Proceedings. Slovak Museum o f Nature Protection and Spe­

leology, The Management o f Slovak Show Caves, Institute of Geography o f the Slovak Academy o f Sciences, Liptovskÿ Mikulaś, Bratislava, pp. 11-15.

Imbrie, J., Hays, J. D., Martinson, D. G., McIntyre, A., Mix, A. C., Morley, J. J., Pisias, N. G., Prell, W. L. & Shackleton, N. J., 1984. The orbital theory of Pleistocene climate: support from a revised chronology of the marine 5 l80 record. In: Berger, A., Imbrie, J., Hays, J., Kukla, G. & Saltzman, B. (eds.), M ilankovitch a nd Climate, Understanding the Response to A stronom ical Forcing. Part I. Reidel, Dordrecht, pp.

269-305.

Ivanovich, M. & Harmon, R. S., 1992. Uranium Series D isequili­

brium. Applications to E nvironm ental Problem s. Clarendon, Oxford, pp. 571.

Jennings, J. N., 1985. K arst Geomorphology. Blackwell, New York, .293 pp.

Jones, B. & Kahle, C. F., 1993. Morphology, relationship, and origin o f fiber and dendritic calcite crystals. J. Sediment.

Petrol., 63: 1018-1031.

Kukla, G., 1977. Pleistocene land - sea correlations I. Europe.

Earth Sei. Rev., 13: 307-374.

Lauritzen, S.-E., 1981. “A ge4U 2U ”. Program fo r reading AD - C A M energy spectra, integration peak-correction a n d calcu­

lation o f 2 ^ T h o r iu m /34Uranium series. Computer Program Turbo Pascal Code, 5,000 lines. Department o f Geology, Bergen University.

Lauritzen, S.-E., 1995. High-resolution paleotemperature proxy record for the last interglaciation based on Norwegian spe­

leothems. Quatern. Res., 43: 133-146.

Palmer, A. N., 1987. Cave levels and their interpretation. Nat.

Speleol. Soc. Bull., 49: 50-66.

Pease, P. P., Gomez, B. & Schmidt, V. A., 1994. Magnetostratigra- phy o f cave sediments, Wyandotte Ridge, Crawford County, Indiana: towards a regional correlation. Geomorphology, 75-81.

Schmidt, V. A., 1982. Magnetostratigraphy o f sediments in Mam­

moth Cave, Kentucky. Science, 217: 827-829.

Schroeder, J. & Ford, D. C., 1983. Clastcic sediments in Castel- guard Cave, Columbia Icefields, Alberta, Canada. Arct. Alp.

Res., 15: 451-461.

Sunagawa, I., 1994. Nucléation, growth and dissolution o f crystals during sedimentogenesis and diagenesis. In: Wolf, K. H. &

Chilingarian, G. V. (eds.), D iagenesis, IV. D evelopm ents o f Sediem entology, 51. Elsevier, Amsterdam, pp. 19—47.

Sweeting, M. J., 1973. Karst Landforms. Columbia University Press, New York, 362 pp.

Swinnerton, A. C., 1932. Origin o f limestone caverns. Bull. Geol.

Soc. Am., 43: 662-693.

Tauxe, L., Deino, A. D., Behrensmayer, A. K. & Potts, R., 1992.

Pinning down the Brunhes-Matuyama and Upper Jaramillo boundaries: A reconciliation of orbital and isotopic time scales. Earth. Planet. Sei. Lett., 109: 561-572.

Thompson, H. P., Ford, D. C. & Schwartz, H. P., 1974. Continen­

tal Pleistocene climatic variation from speleothem age and isotopic data. Science, 184: 893-895.

Tucker, M. E. & Wright, V. P., 1991. Carbonate Sedimentology.

Blackwell, Oxford, 482 pp.

Warwick, G. T., 1976. Geomorphology and caves. In: Ford, T. D.

& Cullingford, C. H. D. (ed.), The Science o f Speleology, Academic Press, London, pp. 61-125.

Webb, J. A., Fabel, D., Finlayson, B. L., Ellaway, M., Li Shu &

Spiertz, H.-P., 1992. Denudation chronology from cave and river terrace levels: the case of the Buchan Karst, southeastern Australia. Geolog. Mag., 129: 307-317.

450

White, W. B., 1988. G eom orphology a n d H ydrology o f K arst Terrains. Oxford University Press, pp. 464.

Williams, P. W., Lyons, R. G., Xunyi, W., Lingchang, F. &

Haosheng, B., 1986. Interpretation o f the paleomagnetism of cave sediments from a karst tower at Guiln. (In Chinese, English summary). Carsologia Sinica, 6: 113-125.

Streszczenie

DATOW ANIE NACIEKÓW Z DEM ENOVSKIEJ LODOW EJ JASKINI ETAPEM W OKREŚLENIU

W IEKU DEM ENOVSKIEGO SYSTEMU JASKINIOW EGO (NIŻNE TATRY, SŁOW ACJA)

Helena Hercman, Pavel Bella, Jerzy Głazek, Michał Gradziński, Stein-Erik Lauritzen & Reidar L0vlie Demenovski System Jaskiniowy (DCS) (Fig. 1) jest powsze­

chnie podawanym w literaturze przykładem wielopoziomowego systemu jaskiniowego. Poszczególne poziomy tego systemu Droppa (1966, 1972) skorelował z poziomami teras rzek powierz­

chniowych - Demenovki, a także Vagu i innych jego dopływów.

Poziomom tym przypisał wiek kolejnych zlodowaceń posługując się klasycznym alpejskim podziałem czwartorzędu (Fig. 2). Deme- novska Lodowa Jaskinia (DIC) jest rozwinięta głównie na IV po­

ziomie jaskiniowym ok. 45 m ponad poziom dzisiejszego dna Demenovki. Droppa (1966) wiąże powstanie tego poziomu ze zlo­

dowaceniem Mindel II.

Próby polew naciekowych do badań laboratoryjnych zostały zebrane w sali Zâvrtovÿ dóm (Fig. 3, 4, 5). Wiek wszystkich polew określono metodą 230Th/234U, a wiek najstarszych polew metodą 234U/238U. Zostało określone także namagnesowanie 11 prób z najstarszej polewy. Wykonane zostały również obserwacje mik­

roskopowe pobranych prób.

^ Wyniki datowań 2 °Th/234U przedstawia Tabela 1 i Fig. 8-10, a - U /'38U Tabela 2 i Fig. 7. Opierając się na wynikach datowa­

nia można wyróżnić cztery generacje nacieków o następującym wieku: ok. 685-410 ka, ok. 170-140 ka, ok. 104—70 ka i młodszym od 5.6 ka. Badane próby najstarszej generacji nacieków, i znajdu­

jącego się bezpośrednio pod nią scementowanego namuliska sili- koklastycznego cechują się normalnym namagnesowaniem (Fig.

6).

Pierwszą i drugą generację nacieków tworzą głównie kolum­

nowe kryształy kalcytu świadczące o dogodnych warunkach wzro­

stu tych generacji. Powierzchnie korozyjne są w nich stosunkowo nieliczne. Natomiast generacja trzecia, występująca ponad nie­

wielkiej miąższości osadami klastycznymi (Fig. 11) zbudowana jest z różnorodnie wykształconych kryształów kalcytu z licznymi powierzchniami nukleacji (Fig. 12) i powierzchniami korozyjny­

mi, co świadczy o niestabilnych warunkach podczas jej wzrostu.

Generacja ta jest oddzielona od nadległej, najmłodszej czwartej generacji czytelną powierzchnia korozyjną.

Poszczególne generacje nacieków jaskiniowych zostały sko­

relowane ze stadiami tlenowymi zapisanymi w osadach głęboko- morskich (cf. Imbrie et al., 1984; Fig. 13): generacja pierwsza ze stadiami 20-11, druga ze stadium 6, trzecia ze stadium 5c-5a, a czwarta ze stadium 1. Problem wzrostu trzeciej generacji nacie­

ków w czasie “zimnego” stadium 6 i ciągły wzrost nacieków od 20 do 11 stadium jest na obecnym etapie badań trudny do wyjaśnie­

nia.

Osady silikoklastyczne zdeponowane pomiędzy datowanymi polewami naciekowymi, powstanie studni w południowej części sali, spełzywanie osadów klastycznych i pokruszenie trzech star­

szych generacji polew było spowodowane dopływem do jaskini inwazyjnych wadycznych wód. Wody te związane były z topnie­

niem kolejnych zlodowaceń.

Otrzymane wyniki datowania nacieków i wyniki badań paleo­

magnetycznych świadczą, że odwodnienie IV poziomu jaskinio­

wego nastąpiło pomiędzy 780 ka i ok. 685 ka. Droppa (1966), na podstawie korelacji z terasami rzek powierzchniowych wiązał powstanie IV poziomu jaskiniowego ze zlodowaceniem Mindel II.

Z porównania estymowanego wieku odwodnienia IV poziomu jas­

kiniowego z przyjmowanym obecnie wiekiem zlodowacenia Min­

del II, który wynosi ok. 500 ka (Kukla, 1977) wynika, że pogląd Droppy (1966) dotyczący wieku IV poziomu jaskiniowego musi zostać zrewidowany. Dowodzi to, że tempo wcinania się Deme- novskiej doliny w młodszym czwartorzędzie było niższe niż do­

tychczas zakładano, a rzeźba Niżnych Tatr jest starsza niż dotych­

czas uważano. Odpowiednio starsze są więc również wyższe (tj.

V-IX) poziomy jaskiniowe DCS.