Overconsolidated sub-till clays in Herlandsdalen, lower Numedal, south Norway

ELEN ROALDSET

Roaldset, E. 1980. Overconsolidated sub-till clays in Herlandsdalen, lower Numedal, south Norway. Norsk Geologisk Tidsskrift, Vol. 60, pp. 39-51. Oslo 1980. ISSN 0029-196X.

A description is given of overconsolidated sub-tiU day sediments situated about 260 m a.s.l., i.e. 90 m above the 1ocal marine limit ofi the last deglaciation. The clay sediments contain neither macro- nor micro-fossils. Textural, mineralogical and geochemical investigations, strongly supported by the Ce deficient lanthanide abundance pattern, indicate marine depositional environment. The clay sediments are assumed to be of Middle-Weichselian age and are thus correlated with the Sandnes Interstadial, described from SW-Norway, and its analogues.

E. Roaldset, Institutt for geologi, Universitetet i Oslo, Boks 1047, Blindern, Oslo 3, Norway.

In the Numedal river basin sub-till sediments have previously been described from the north­em parts of the area (Roaldset 1973a, Rosenqvist 1973). This paper presents data on the ftrst sub­till clays in the lower part of the Numedal river basin, where in 1973 in the valley Herlandsdalen, overconsolidated sub-till clays situated 260 m a.s.l. (Fig. l) were discovered.

Geological setting A sketch map of the superficial deposits, given in Fig. 2, shows thick till cover in the SW, and thin discontinuous cover of till NE of Herlandsda­len.Dead ice terrain is locally developed, and at K viberg-Olsli a glaciofluvial delta was deposited and built up to the postglacial marine limit (173± 2 m a.s.l.; KorbØl 1972), simultaneously with deposition of marine clays in the ancient 'Numedalsfjorden'. Accordingly the sub-till clays are found almost 90 metres above the post-glacial marine limit.

The main directions of glacial striae, as observed north and southeast of the Herlandsda­len area, are N 100c and N 160c - 110c respec­tively (Holtedahl & Andersen 1960, D.E. Nilsen pers. comm. 1978, K. Pederstad, pers. comm. 1978). Like the valley innfills of till in valleys tributary to Gudbrandsdalen (Mangerud 1965, Garnes & Bergersen 1977), the preservation of till in SW-areas of Herlandsdalen may be ex­plained by their lee-side position during ice flow obliquely across the valley (Hillefors 1973).

The bedrock consists of Permian monzonite and alkali granite of the Oslo lgneous Province. Precambrian rocks occur about lO km WNW of Herlandsdalen (BrØgger & Schetelig 1923).

The Herlandsdalen sediments The sub-till clay was exposed near the head of Herlandsdalen (Fig. 2). The exposed sections are shown in Fig. 3.

The Locality l sediments are glacially tectonized and display small fissures and faults. Well rounded pebbles of Precambrian rocks are scat­tered throughout the clay bed, becoming more abundant upwards. A 30 cm thick bed of silty sand succeeds the ela y, with distinct contact to the bluish grey zone above. This zone, being a mixture of clay and till material, gradually changes over to a yellowish brown till. Frag­ments of rock as well as of consolidated clay occur in the til!.

The Locality 2 sediments consist of homogenous fine grained clay in distinct contact with alkali granite below, and till above. The ela y shows no sign of glacial tectonism or disturbance, and is presumably preserved in situ. The till is fur­rowed by small channels (5-20 m wide, 2-10 m deep). The coarser till fragments consist of well rounded cobbles of Precambrian rocks. Frag­ments of Lower Palaeozoic or of Permian rocks were not identified in the coarsest fractions, but occur in small amounts, below 10 per cent in total, in the grave! fraction.

40 E. Roaldset NORSK GEOLOGISK TIDSSKRIFT l (1980)

o

o o

ls - SKI 19800- 9900 BP l �

RA 110200 BP)

20km

lOO m.a.s.l.

Fig. l. Oslofjorden and the location of the Herlandsdalen area. The 300m contour is indicated. Terminal moraines from the last

deglaciation are indicated with broad black lines.

Fig. 2. Superficial deposits in the Herlandsdalen area (KorbØII972, unpublished data, Numedalsprosjektet, Institutt for geologi,

Universitetet i Oslo).

NORSK GEOLOGISK TIDSSKRIFT l (1980)

HERLANDSDALEN

SOOm

Sub-ti/1 clays in Numedal 41

Bedrock with small areas eovered by lill

Thin cover with tiU

Thick cover with un

Glaciofluvial gravet and sand

Fluviat sand

Bog

- Eskor

Orainag.e ehannel

Glaciofluvial terrace

SOm

42 E. Roaldset

M o

2

4

6

Loe. 1.

* 7Tf////J/////////////77;

Ablotion lill

Till

Till/Cloy

Sand

Cl a y

Slumped mot•rial

M o

2

4

6

Loe. 2.

NORSK GEOLOGISK TIDSSKRIFT l (1980)

lill

Silt y clay

not excavat•d ?r!!!!!!!!� Alko li gronite 8 +++++++++++++++++

+++++++++++++++++

Fig. 3. Sections through the till and clay sediments. The clays are bluish grey and the tills yellowish brown. Asterisks show the

sample positions. At Loe. l (disturbed) the clay in the upper parts contains well-rounded pebbles ofPrecambrian rocks. The sand

bed rna y represent a fluvial or glaciofluvial event.

Neither macro- nor microfossils were detected in clay samples prepared for pollen, foramini­fera, and diatom analysis.

Sediment texture and particle size distribu­tions are illustrated by means of electron micrographs in Fig. 4. The Loe. 2 clay shows a great abundance of <2 �-tm materials with a few scattered coarse fragments, while coarser frag­ments dominate in the Loe. l ela y.

Experimental methods pH measurements were carried out in situ, using a Beckman H5 pH-meter and a combined glass­calomel electrode.

The clay samples were examined without any pretreatment with a JEOL scanning electron microscope.

Geotechnical parameters were determined by conventional methods according to Skempton (1953).

Material coarser than 63 �-tm was wet sieved employing screens of whole phi unit intervals, while the grain size distribution of the fine frac­tion was determined by pipette analysis (Krum­bein & Pettijohn 1938).

Mineralogical analyses of the clay-sized frac­tion (finer than 2 �-tm) were made with oriented samples by X-ray diffractometry (XRD). The

methods of mineral identification were essen­tially based on Brown (1961).

The major elements were determined by X-ray fluorescence spectrography (XRF). Ferrous iron was determined by dissolving the sample powder in cold concentrated HF with subsequent col­orimetric titration against K2Cr207• The igni­tion loss was determined as the weight loss of material dried at ll0°C after heating for 2 hours at 1050°C.

The relative distributions of lanthanide (Ln) elements were determined by a modified GEC/ AEI MS 702 solid source mass spectrometer (Taylor 1965, Nicholls et al. 1967 ). A description of the analytical procedure is given by Prestvik & Roaldset (1978). Christie (1977) showed that the precision of the method is 8 per cent, the accuracy within sample is 10 per cent and the accuracy of the method, which is a measure of reproducibility of sample preparation, homogeneity and precision combined, is 25 per cent.

Results and comparisons pH-measurements In sit u the pH value was 7 . 5 for the sub-till clays, which is of the same order as for Scandinavian postglacial marine clays isostatically lifted above

NORSK GEOLOGISK TIDSSKRIFT l (1980) Sub-till c/ays in Numedal 43

Fig. 4. Sub-tiU clays. Scanning electron micrographs of horizontal sections: Loe l. A- 300X, B- 3000x; Loe 2. C- 300x,

D-3000x.

sea-level (J. Moum, pers. comm. 1973), and higher than normally found for Numedal tills (unpublished data, Numedalsprosjektet, Institutt for geologi, Universitetet i Oslo).

Determination of consolidation stress

The sub-till clays had a considerable hardness and did not crack during drying. They were

therefore assumed to be greatly overconsoli­dated. The clays did not shrink with further drying, which proved that they have been con­solidated to beyond the shrinkage limit, which is normally 15-20 relative per cent beyond the plastic limit (Hogentogler 1937). The following considerations and calculations are analogous to those made by Rosenqvist (1973) for the

FønnebØfjord sub-till clay in upper Numedal.

44 E. Roaldset

Table l. Geotechnical parameters for � sub-till clay (Loe. 2, undisturbed).

Water content (WJ P1astic limit (WP) Liquid limit (�\') Void ratio (e) P1asticity index (PI = WL - WP)

COntent of Clay < 2 lJm (c) ACtivity (A = Pl/C)

14.9 ' 17.3 ' 32.3 '

o. 40

15. o '

55. o ' o. 27

l)Calculated on the basis of a specific qravity of the particles of 2. 7 g/cm3.

The fundamental geoteehnical data for the in situ sub-till clay (Loe. 2) are given in Table l. By using Skempton's (1953) empirieal eurves at liquid limit of 50, plasticity index 25, and a void ratio of 0. 40, the data correspond to an approxi­mate depth of 2000 ft. in normally consolidated sediments. Similar figures were obtained by ex­trapolation of data for Horten ela y (from Skemp­ton 1953), while such consolidation stresses Iie outside Bjerrum & Rosenqvist's (1958) experi­mental range for Åsrum ela y, lower Numedal. Applying the formula given by Skempton (1953) for the relationship between void ratio and effec­tive overburden pressure, the effective load on the Loe. 2 sub-till clay must have been l03·86N per m2 (or 7 40 tons/m2). This is slightly lower

99.5 99 98 97 95

90

NORSK GEOLOGISK TIDSSKRIFT l (1980)

than the estimated load of 1000 tons/m2 for FønnebØ clays, upper Numedal (Rosenqvist 1973).

Obviously the eonsolidation stress for the Herlandsdalen sub-till clay was much higher than the weight of the overlying till, and it is unlikely that this material alone eonsolidated the sediments. Thus the consolidation is due to an overlying glacier with a thiekness of at least 7--800 m. This must correspond to a major glacial advance.

Grain-size characteristics The grain size distribution was plotted as weight per cent (log probability scale) versus phi (<l>) whieh is defined as - log 2 particle diameter in millimetres (Fig. 5). In Fig. 5 the grain size increases from left to right whieh is the opposite of Folk & Ward's (1957) proposal. Consequently the equations for particle distribution must be modified (KorbØl 1972). The grain size data in Table 2 gives the following properties of the samples investigated (Folk 1968).

Loe. l. Si/ty clay, very poorly sorted, strongly fine skewed mesokurtie. Sand, poorly sorted, eoarse skewed very platykurtic. Til/, very poorly sorted, fine skewed leptokurtic.

" ,, . 11 :

; l : , l :

... �' ! 95 .,"',... ' :

80 70 60 50

.... ... .... .... Ti li C loc.t) ........ ,' '\.. ."."' .""' 84

40 r 30 20

10

1 0.5

��.... ... ..··· .........

,��-"'.., ... ········ ,f' .··'\

.... :,

...

".""'"'

/' / Sand ( loc.l)

.," ,. "', .... , "" ,. "' ... ,.-' ,' Till '<loc.2)

." ,.". .·

:.�� .. �.� ...................... ··

O.l L__._ _ _.___. _ _.__..___._ _ _.___. _ _.__..___.__...___, _ __.__..__� •11 •10 •9 •8 •7 •6 •5 •4 •3 •2 •1 o -1 -2 -3 -4

0- units

75

16

Fig. 5. Cumulative weight per

cent (log probability scale) plot·

ted against grain size in phi (<11=­

log,d).

NORSK GEOLOGISK TIDSSKRIFT l (1980) Sub-till clays in Numedal 45

Table 2. Sorne standard size and particle distribution parameters (Folk 1 Ward 1957) for

the Herlandsdalen clay and till.

Loe. l Loe. 2 Measure Sub-til l Sand Till Sub-til l Til l

ela y ela y

Graphic rnean l) 7.3 0.16 2.8 10.0 2.3

sorting (Inclusive graphic 2) standard deviation)

3.4 1.2 3.0 2.9 2.7

Skewness (Inclusive31raphic skewness)

0.59 - 0.18 0.14 - 0.08 - 0.08

Graphic kurtosis4) (Peakedness)

1.00 0.62 1.14 0.96 l. 21

l)

2)

11 "' z

01 : 4>16 - 4>8.4 + <P5 - 4>9s

4 6.6

3) Sk1

4) KG

Loe. 2. Clay, very poorly sorted, near sym­metrical mesokurtic. Till, very poorly sorted, near symmetri­cal leptokurtic.

The in situ clay is finer than most marine clays around Svarstad (KorbØI 1972), and similar to the most fine-grained marine clays in the Skager­rak (RØnningsland 1976), and sub-till lacustrine clays of central South Norway (Rosenqvist 1973, Vorren & Roaldset 1977). The tills have grain size distributions typieal for basal tills of the surrounding area (KorbØI1972).

By plotting sorting (u1) versus mean grain size (Mz) for late- and postglacial sediments of the Svarstad area, a clear distinction appeared be­tween marine clays, tills, fluvial sand, and grave! (KorbØl 1972). The Herlandsdalen samples fit this division. The overconsolidated clays and the sand plot within the areas for marine clays, glacio-fluvial sediments, and fluvial sediments respectively (Fig. 6).

Clay mineralogy The clay fractions (< 2 J.tm) of the tills and the sand bed gave almost identical X-ray

=

�16 + �84 - 2�50 + 4>s + �95 -2�50

21<Pl6 - �84) 2 <�s - �95l

<Ps - 4>95 2.44 (4>25 - 4>75)

diffractograms. The clay beds differed from the eoarser sediments but were similar to each other. For this reason only the X-ray diffraetograms for the Loe. 2 clay and till are presented (Fig. 7).

The phyllosilicates present in the clay frac­tions of the tills and the sand bed are illite (lOÅ), vermiculite, and a mixed layer illite-vermiculite (l0-14Å). In addition mierocline (3.25Å), plagioclase (3.19Å), amphibole (8.4Å), quartz (4.26Å), and small amounts of smectite were detected. After treatment with HCl the 7 Å peak disappears, thus eliminating the presenee of ka­olinite.

The sub-til! clay eonsists of illite, chlorite (l4Å) together with traces of smectite, and mixed layer illite-chlorite minerals. Besides th­ese, amphibole, quartz, mierocline, and plagioclase are present. Not in this material either was a 7 Å kaolinite peak observered after HCI treatment. The X-ray diffraetograms for this clay resemble the diagrams obtained for river mud and some till samples from Numedal (Roaldset 1972) as well as marine Skagerrak clays (RØnningsland 1976).

Treatment of the < 2 JL m material with sodium-tetraphenyl-boron (NaTBP) revealed in

46 E. Roaldset

4 MARINE / e2 SEDIMENTS l

l

/

3 .1 / TILLS

-- -

l l

l l ---

1 ------

__ , ___ - ---

l l l

01

l

l l l

NORSK GEOLOGISK TIDSSKRIFT l (1980)

o lill

o Glaciofluvial (?) • Sub-til l ela y 1 Loc.1

2 Loe 2

GLACIO- FLUVIAL AND FLUVIAL SEOIMENTS Fig. 6. Plot of sorting (inclusive

graphic standard deviation) ver­sus graphic mean. The divisions were obtained for different sedi­ment types from the lower Herlandsdalen - Svarstad region (after KorbØI 1972).

8 6 2p

3 331

Sub- lill ela y C fraction < 2 �)

4 63 fl

2 o 1mm

Effective diameter

71 841 10A 141

Glycol

4so•c

-2 o Mz

3 JJl

Ti li ( fr•ction

Fig. 7. X-ray diffractograms of the clay fraction (<2p.m) of the overconsolidated clay and the overlying till (Loe. 2).

both cases the illites to be of di- and trioctahedral type (DeMumbrum 1%3, Robert 1973, Schmith & Scott 1973, Rueslåtten 1976).

The Jack of chlorite in the till and its appear­ance in the underlying clay samples and the

opposite distribution of vermiculite deserves at­tention. According to Garrels & Christ (1965) chorite is stable in marine environments, but unstable in weakly acidic ones. The high content of vermiculite minerals in the till can be attri-

NORSK GEOLOGISK TIDSSKRIFT l (1980) Sub-till clays in Numedal 47

Tab1e 3. loiajor chemical composi ti on (weight per cent) for tbe c1ay

fr action (<2 um) of the Herlandsdalen sediments.

Loe. l Loe. 2 Elements Sub till Sand Till Sub-till Till

c1ay bed c1ay

5102 49.6 56.2 63.3 53.2 54.9

Ti02 l. 07 0.95 0.95 1.42 0.86

A12o3 15.7 16.5 15.2 16.7 15.9

Fe2o

3 10.7 7, 2·6 4.93 6.67 4. 54

FeO 3.10 2.80 2.54 3.12 2.50

Mn O 0.12 0.13 0.12 0.12

Mg O 5.15 3.08 2.40 4.13 l. 98

ca o 2.52 3.00 3.14 2.68 2.79

Na2o l. 59 2.41 2.53 2.04 2.18

K20 3.75 3.66 3.12 4.34 2.49

P2

05

0.19 0.14 0.23 0.18 0.26

Loss ignition 5.50 3.96 2.67 5.47 12.3

Total 99.0 100.1 101.1 100.0 100.8

buted to formation of vermiculite from biotite and plagioclase (Meunier & Velde 1976, 1977) as well as to transformation of chlorite and trioctahedral illite during subaerial weathering processes.

The mineralogical analyses do not identify the sub-till clay in Herlandsdalen as undisputable manne.

Chemical composition The chemical composition of the< 2�.tm fraction of the ela y and till is given in Tab1e 3. The data are in agreement with chemical data of the clay fractions of Numedal sediments and when re­calculated to mol percentages of basic oxides plot within the same area as Numedal clays in a (Fe203 + FeO + Mg0)-Al203-(K20 + Na20 + K20) triangle (Roaldset 1972).

Roaldset (1972) found a consistent geochemi­cal variation for mol ratios MgO/ Al203 and K20/ Al203 between marine and non-marine clays. The marine clays usually had higher and the non-marine clays lower, mol ratios than MgO/ Al203 = 60 per cent and K20/ Al203 = 25 per cent. U sing these criteria, the Herlandsdalen sub-till clays (MgO/ Al203 = 64 and 87 per cent, K20/ Al203 = 29 and 28 per cent) fall within the marine region, while the tills and sand plot in the non-marine group. Analyses of lacustrine sub-till

clay beds and laminae from Hardangervidda and upper Numedalen gave mol ratios Mg0/Al203 = 50-60 per cent, except for one value of 70 per cent. Thus, although the major chemical com­position would seem to indicate a marine origin for the ela y, a lacustrine depositional environ­ment cannot be completely ruled out.

Lanthanide (Ln) elements The distributions of the Ln elements within the

< 2 �.tm fraction, relative to La= 100, are given in Table4.

To facilitate their presentation it is usual to divide one Ln distribution, element by element, by another distribution, plotting the resulting ratios in a logarithmic scale against a linear scale of atomic number (Masuda 1962, Coryell et al. 1963). The data for the present work were thus normalized to their average compositions and plotted together with a true marine clay from the Skagerrak in Fig. 8.

The lanthanide distribution pattern may be used as an environment indicator within certain limits (cf. Balashov et al. 1964, Ronov et al. 1967, Fleischer & Altschuler 1969, Martin et al. 1976). The pH-dependent oxidation of cerium is well known. At pH 8.2, which corresponds to marine conditions, nearly all cerium exists as CeH, while the prevailing ion at. pH 7 is Ce3+.

48 E. Roaldset NORSK GEOLOGISK TIDSSKRIFT l (1980)

Table 4. Relative lanthanide (Ln) distributions (I.a = 100) for the

Her1andsdalen eamplee.

Loc.l Loe. 2 silb-Hn Sand silb-tlli

Element c1ay bed Til l clay Til1

La 100 100 100 10 0 100

Ce 215 224 321 207 916

Pr 44 29 9.4

Nd 152 86 22

Sm 50 20 4.2

Eu 15.5 2.2 0.5

Gd 98 27 5.4

Tb 10.4 1.8 o. 54

Dy 40.5 8.6 4.0

Ho 8.2 1.9 0.75

Er 29 9.6 3.2

Tm 3.3 0.27

Yb 17.7 7.7 l.S L u 5.5 0.26

The oxidation of ca+ to Ce4+ explains the cerium enrichment in hydrolysate minerals and sub­aerial weathering products, and the cerium de­fiency in ocean water (Goldberg 1961, Carpenter & Grant 1969, Piper 1974 a,b). According to Hogdahl (pers. comm. 1967) the North Atlantic deep water has a pronounced cerium deficiency.

Martin et al. (1976) investigated the behaviour

16.4 29 65 81

13.5 31

1.6 5.3

16.9 40

1.7 4.4

12.5 34

2.5 5.7

6.4 19.0

o. 74

4.4 13.1

0.92

of the Ln's in the river/ocean boundary in the Gironde estuary, France. For soluble river loads they found a gradual decrease in the Ce/La ration from 1.66 to 0.35 as the salinity increased from O.l %o to 35.5 %o.

The Loe. 2 samples (Fig. 8), the clay and the till above, exhibit pronounced positive and nega­tive Ce anomalies respectively, and thus attach

Loe. l. Gl.acial t�ctonized and redeposited s�diments Loe. 2. Undisturbed sediment �q��n:e lJ • Ill .. Q. E � 0.5

""" / .......... /·\ .. "--/·-· \ ·-· . .,-;- !:-:.·�-.. ::.- .! --------------- -·---' \ . /. . . '\. .

Glaclofluvlal bed • •-• .....-

SmEu Gd Tb Dy Ho Er Tm Yb Lu

• 3

!:>;,:���<�����::::::� r \ • Sub-til! clay :; _....-- ···-- •-•-. m.arint?" • •

/ ..... ...-•-. J \. ·-· ........ 105 Il ·----- . ., ... - ---------- - ------------

\ •/ ',, Skagerak clay / ', (532) •....._ • Marine/ . ........ _.,

........... __ ._.....

La C� Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb L u

Lanthanide atomic numb�r

Fig. 8. Lanthanide comparison diagram relative to the average of the Herlandsdalen samples, together with sample S 32, a true marine clay from Skagerrak (Roaldset & Christie 1975).

NORSK GEOLOGISK TIDSSKRIFT l (1980)

significance when trying to arrive at an under­standing of the depositional conditions.

For the overlying til! the Ce enrichment pro­duces evidence for a high content of subaerial weathering products.

A Ce depletion occurs rarely in postglacial Numedalen-sediments deposited in brackish and non-marine environments, and has been found only in one sample, which was actually from Herlandsdalen (Roaldset 1973b).

The Herlandsdalen ela y, due to the Ce defiency and its resemblance with Skagerrak ela y, may have been deposited in a true marine environ­ment. The possibility of a marine origin requires further discussion as the locality Iies 90 m above the marine limit of the last glaciation.

Discussion and correlation .

The investigations led to the following conclu­sions:

The pH, grain size, mineralogical and major chemical analyses all suggest a marine deposi­tional environment, though Iacustrine conditions cannot be completely ruled out.

The Ce-defiency exhibited in the distribution pattern points to a marine origin of the sub-til! clays.

The degree of overconsolidation of the undis­turbed clay indicates a major glacial advance across the Herlandsdalen area after the clay deposited. Hence the clay was at !east deposited prior to the Weichsel III glacial advance (Lund­qvist 1974), which culminated 17,900 ± 1300 years B.P. (Pisias 1976, see also CLIMAP pro­ject members 1976, Lawrence Gates 1976, Porter et al. 1977).

As the sub-til! clay sediments Iie 260 metres above present-day sea-level, i.e. about 90 metres above the postglacial marine limit at this locality, a marine depositional milieu cannot be explained by isostatic/eustatic conditions comparable with those which have prevailed since the last de­glaciation.

The Jack of organic material, pollen diatoms, and foraminifera unfortunately precludes age datings by means of biostratigraphical and radiocarbon methods.

Age and correlation

The til/. - The thin ablation til! on top of the basal til! (Fig. 3, Loe. l) was probably deposited

4- Geologisk Tidsskr. 1/80

Sub-till clays in Numedal 49

during the last deglaciation (Preboreal?) about 10,000 years ago. The basal til! may represent either the W eichsel Il or Ill advance (Lundqvist 1974). Extensive studies of distribution and genesis of tills in central south Norway signify that the basal tills were most probably deposited during the onset of a glaciation (Bergersen & Garnes 1977).

The clay. - Since evidence exists that the Ee­mian shoreline in SW Norway and SW Sweden was only about 10-20 metres above present sea-leve! (Mangerud 1970, Feyling-Hanssen 1971, Hillefors 1974), and since clays of pre-Ee­mian age have not been found in Norway, the clays in Herlandsdalen may be assumed to be younger than Eemian. Their stratigraphical posi­tion and overconsolidation appear to leave open the possibility that the ela y was deposited during a Middle Weichselian interstadial prior to the great Weichsel advance.

Taking into consideration the proposed marine origin, the clay must have been deposited at a time when the sea leve! was about 100 m above the local postglacial marine limit. The cause of such high sea-leve! has been generally attributed to glacial isostasy. For the Weichsel glaciation, Asseev (1968) considers that the maximum isostatic depression was about 1000 m, and Nansen (1928) and Gutenberg (1941) give minimum values of 550-700 m. As no eustatic rise in sea level above present niveau has occur­red within the last 30,000-35,000 years (Milliman & Emery 1968), glacio-isostatic depression alone would explain a marine transgression reaching 250-300 m above present sea leve!.

Recent investigations present evidence of large postglacial vertical displacements (up to 100 m) within the southern Fennoscandian Bor­der zone and the Baltic Shield (Lagerlund 1977 a,b, Lundqvist & Lagerbiick 1976, Lagerbiick 1977, Morner 1977 a,b). Worthy of mention is also the observation by Reusch (190 l) of vertical displacement across glacial striations at Gryte­fjellet, near Vøringsfossen, west Norway. Thus, referring to these investigations a tectonic uplift of the Herlandsdalen area cannot be excluded.

From W and SW Norway marine clays lying far (at Jæren up to 250 m) above the postglacial shoreline has been described by several authors (Grimnes 1910, Kaldhol 1930, 1931, Feyling­Hanssen 1966, 1971, 1974, Garnes 1976). At Foss-Eigeland, near Sandnes, an almost com­plete W eichselian profile has been recorded

50 E. Roaldset

(Bergersen & Pollestad 1971, Feyling-Hanssen 1974). Here, below a 3-5 m thick basal till, a clay zone occurs with its base 65 m above present day sea leve!. The clay represents the upper parts of a Middle Weichselian fining upwards sediment sequence corresponding to a marine transgression before the last large glacial advance. This transgression has also been rec­ognized in the Egersund district (Garnes 1976).

The Herlandsdalen clay most probably corre­lates with the clay unit B) at Foss-Eigeland which according to Feyling-Hanssen (1971, 1974) was deposited during the Sandnes Interstadial (50,000-20,000 years B.P.). This Middle Weich­selian interstadial is correlated with the Hirtshals Older Yoldia Clay (Andersen 1971), the Port­

landia arctica zone of the Skærumhede I (Jessen et al. 1910), and the zone I of the Skærumhede Il sequence (Bahnson et al. 1973), the Swedish analogues the Gotaelv and Y ounger Dosebacka­Ellesbo interstadials (Brotzen 1961, Hillefors 1969, 1971), and the Cape Broughton Interstadial (Mid-Wisconsian) known from Arctic Canada (Feyling-Hanssen 1976a,b).

Conclusions

The lanthanide abundance pattern may point to a marine origin for the sub-til! clay sediments in Herlandsdalen. Their degree of overconsolida­tion and the fact that they Iie about 90 m above the postglacial marine limit support a correlation with the Sandnes Interstadial. Thus in the terms of Mangerud et al. (1974 ), the ela y was deposited during a Midd le W eichselian Interstadial. The interstadial conditions were followed by the great Weichsel Il (?)/Ill advance(s) (Lundqvist 1974). Glaciers from the NW overrode the clay (Loe. 2), eroding it and transporting part of it as thrust slices and also incorporating some clay in the til! (Loe. l) which was subsequently depo­sited.

The author wishes to keep open the question whether the high Middle Weichselian sea leve! in Herlandsdalen was due to glacio-isostatic depression, or to neotectonic movements, or combination of such effects.

Acknowledgements. - BjØrg E. Alfsen, Kari Henningsmoen, Magne LØfaldli, and Tor M. Rønningsland assisted during the analytical work, Toril and Eigil Whist made most of the final drawings, and Adrian Read corrected the English text. The work was stimulated by discussions with Olav H.J. Cbristie, Per Jorgensen, and Ivan Th. Rosenqvist. To all these persons I

NORSK GEOLOGISK TIDSSKRIFT l (1980)

offer my sincere thanks. The financial support of Norges Almenvitenskapelige Forskningsråd grant D. 40.31-11 is grate­

fuUy acknowledged.

References Andersen, A.-L.L. 1971: Foraminifera from the Older Yoldia

Clay at Hirtshals, 159-184. In Feyling-Hanssen, R. W., Jorgensen, J.A., Knudsen, K.L. & Andersen, A.-L.L., Late Quaternary Foraminifera from Vendsyssel, Denmark and Sandnes, Norway. Bull. Geo/. Soc. Den. 21.

Aseev, A.A. 1968: Dynamik und geomorphologische Wirkung

der europaischen Eisschilde. Petermann' s Geo gr. Mitt ei/un­gen, IJ2, 113-II5.

Bahnson, H., Petersen, K. Strand, Konradi, P. B. & Knudsen, K.L. 1974: Lithology and biostratigraphy of Quaternary deposits of the Skærumhede Il boring. Dan. Geo/. Unders. Årb. 1973, 27-f>2.

Balashov, Yu.A., Ronov, A.B., Migdisov, A.A. & Turans­kaya, N.V. l %4: The effect of climate and facies environ­ment in the fractionation of the rare earths during sedimen­tation. Geochem. lnt. JO, 951-969.

Bergersen, O.F. & Pollestad, B.A. 1971: Evidence of fossil

ice wedges in early Weichselian deposits at Foss-Eikjeland, Jæren, south west Norway. Nor. Geogr. Tidsskr. 25, 39-45.

Bergersen, O.F. & Garnes, K. 1971: Evidence of sub-till sediments from a Weichselian interstadial in the Gudbrandsdalen valley, central east Norway. Nor. Geogr. Tidsskr. 25, 99-108.

Bjerrum, L. & Rosenqvist, I.Th. 1956: Some experiments with artificially cemented clays. Geotechnique 6, 1-13.

Brotzen, F. 1961: An lnterstadial (radiocarbon dated) and the sub-stages of the last glaciation in Sweden. Geo/. Foren. Forh. Sth. 83, 144-150.

Brown, G. (ed.) 1961: The X-ray ldentification and Crystal

Structures of Cia y Minerals. Mineralogical Society, London.

Brogger, W.C. & Schetelig, J. 1923: Geologisk oversiktskart over Kristianiafeltet. Nor. Geo/. Unders.

Carpenter, J.H. & Grant, V.E. 1969: Concentration and state of cerium in coastal waters. J. Mariner Res. 25, 228-238.

Christie, O.H.J. 1977: Ana/ytical Procedure for the Determi· nation of Lanthanoids in Six African Amphibolites. Progress

Report series, Massespektrometrisk laboratorium, Uni­versitetet i Oslo.

CLIMAP project Members, 1976: The surface of the lee-Age Earth. Science 191, 1131-1138.

Coryell, C.D., Chase, J.W. and Winchester, J.W. 1963: A procedure for geochemical interpretation of terrestrial rare­earth abundance pattern. J. Geophys. Res. 68, 559-566.

De Mumbrum, L.E. 1963: X-ray diffraction changes by K-loss

from orthoklase and microcline feldspars. Soil Sei. 96, 428-430.

Feyling-Hanssen, R.W. 1966: Geologiske observasjoner i Sandnes-området. Nor. Geo/. Unders. 242, 26-43.

Feyling-Hanssen, R.W. 1971: Weichselian Interstadial Foraminifera from the Sandnes-Jæren area, 72-116. In Feyl­ing-Hanssen, R.W., JØrgensen, J.A., Knudsen, K.L. & Andersen, A.-L.L., Late Quaternary foraminifera from Vendsyssel, Denmark and Sandnes, Norway. Bull. Geo/. Soc. Den. 21.

Feyling-Hanssen, R.W. 1974: The Weichselian section of

Foss-Eigeland, south-west Norway. Geo/. Foren. Sth. Forh.96, 341-353.

Feyling-Hanssen, R.W. 1976a: A Mid-Wisconsian interstadial

NORSK GEOLOGISK TIDSSKRIFT l (1980)

on Broughton Island, Arctic Canada, and its foraminifera. Arctic and Alpine Res. 8, 161-182.

Feyling-Hanssen, R. W. 1976b: The stratigraphy of the Quater­nary Clyde Foreland Formation, Baffin Island, illustrated by

the distribution of benthic foraminifera. Boreas 5, 77-94.

Fleischer, M. & Altschuler, Z.S. 1969: The relationship of the rare-earth composition of minerals to geologic environment. Geochim. Cosmochim. Acta 33, 725-732.

Folk, R.L. 1968: Petrology of Sedimentary Rocks. HemphiU's Drawer M. University Station, Austin, Texas.

Folk, R.L. & Ward, W.C. 1957: Brazos River bar: A study of the significance of grain size parameters. J. Sed. Petro/. 27, 3-26.

Garnes, K. 1976: Stratigraphy and morphogenesis of drup�lins on EigerØya, Rogaland, SW Norway. Arkeol. mus. Stavanger, Skr. l, 53 pp.

Garnes, K. & Bergersen, O.F. 1977: Distribution and genesis

of tills in central south Norway. Boreas 6, 135-147. Garrels, R.M. & Christ, C.L. 1965: So/utions, Minerals, and

Equilibria. Harper & Row, New York. Grimnæs, A. 1910: Jæderens Jordbund. Nor. Geo/. Unders.

52, 1-104. Goldberg, E.O. 1961: Chemistry of the oceans.

Oceanography, Am. Ass. Advanc. Sei., Washington D. C., 583-597.

Gutenberg, B. 1941: Changes in sea lev el, postglacial uplift and

mobility of the earth's interior. Bull. Geo/. Soc. Am. 52. Hammond, L.A. 1976: Pa1eoclimate: lee Age Earth was cool

and dry. Science 191, 455.

Hillefors, Å. 1969: Vastsveriges glaciala historia och

morfologi. Naturgeografiska studier. Medd. Lunds Univ. Geogr. Inst. Avhandl. 31, 407 pp.

Hillefors, Å. 1973: The stratigraphy and genesis of stoss- and lee-side moraines. Bull. Geo/. Inst. Univ. Upps. 5, 139-154.

Hillefors, Å. 1974: The stratigraphy and genesis of the Dosebacka and Ellesbo drumlins. A contribution to the knowledge of the Weichsel glacial history in Sweden. Geo/.

Foren. Stockh. Forh. 96, 355-374. Hogentogler, C.A. 1937: Engineering Properites of Soil. New

York.

Holtedahl, O. & Andersen, B.G. 1960: Glacial map of Nor­

way. Nor. Geo/. Unders. 208. Jessen, A., Milthers, V., Nordmann, V., Hartz, N. & Hes­

selbo, A. 1910: En boring gennem de kvartære lag ved Skærumhede. Dan. Geo/. Unders. 225, 1-175.

Kaldhol, H. 1930: Sunnmores kvartærgeologi. Nor. Geogr. Tidsskr. ll, 1-194.

Kaldhol, H. 1931: Har vi spor efter flere istider i Norge. Nor. Geo/. Tidsskr. 12, 391-448.

Korbol, B. 1972: En kvartærgeologisk og sediment­petrografisk undersokelse omkring Svarstad. Unpubl. cand. real. thesis, Universitetet i Oslo, 78 pp.

Krumbein, W.C. & Pettijohn, F.J. 1938: Manual of Sedimentary Petrography. Appleton-Century-Crofts, Inc., New York.

Lagerback, R. 1977: Unga rorelser i svenska urberget. Forsk­ning och Framsteg 2, 7-14.

Lagerlund, E. 1977a: Forutsattningar for moranstratigrafiska undersokningar på Kulien i nordvast Skåne - teoriutveckling och neotektonik. Dr. thesis, Dept. of Quaternary Geology, University of Lund, Sweden.

Lagerlund, E. 1977b: Till stratigraphy and neotectonism in southern Sweden. Boreas 6, 159-166.

Lawrence Gates, W. 1976: Modelling the lee-Age Climate. Scince 191, 1138-1144.

Lundqvist, J. 1974: Outlines of the Weichsel Glacial in Swe­den. Geo/. Foren. Stockh. Forh. 96, 327-339.

Sub-till clays in Numedal 51

Lundqvist, J. & Lagerback, R. 1976: The Parve Fault: A late

glacial fault in the Precambrian of Swedish Lapland. Geo/.

Foren. Stockh. Forh. 98, 45-51. Mangerud, J. 1965: Dalfyllinger i noen sidedaler til

Gudbrandsdalen, med bemerkninger om norske mammut­

funn. Nor. Geo/. Tidsskr. 50, 167-181. Mangerud, J. 1970: Interglacial sediments at Fjosanger, near

Bergen, with the first Eemian pollen-spectra from Norway. Nor. Geo/. Tidsskr. 50, 167-181.

Mangerud, J., Andersen, S.T., Berglund, B.E., & Donner, J. 1974: Quaternary stratigraphy of Norden, a proposal for terminology and classification. Boreas 3, 109-128.

Martin, J�M., Hogdal, O. & Philippot, J.C. 1976: Rare earth element supply to the ocean. J. Geophys. Res. 8/, 3119-3124.

Masuda, A. 1962: Regularities in variation of relative abundances of lanthanide elements and an attempt to analyse separation-index patterns of some minerals. J. Earth Sei. Nagoya Univ. JO, 173-187.

Meunier, A. & Velde, B. 1976: Mineral reactions at grain contacts in early stages of granite weathering. C/ay Miner. Il' 23 5-240.

Meunier, A. & Velde, B. 1977: Authigenic mineral facies formed during the early stages of granite weathering, 126-128. In Rosenqvist, I.Th. (ed.), Proceedings The Third European Clay Conference, Oslo.

Milliman, J.D. & Emery, K.O. 1968: Sea levels during the past 35,000 years. Science 162, 1121-1123.

Morner, N.-A. 1977a: Past and present uplift in Sweden: glacial isostasy tectonism and bedrock influence. Geo/. Foren. Stockh. Forh. 99. 48-54.

Morner. N.-A. 1977b: I Dyring: Nya ron om landhojning. Forskning och Framsteg l, 32-37.

Nansen, F. 1928: The earth's crust, surface-forms, and isosta­tic adjustment. Nor. Vid. -Akad. Oslo Avh. 1927 l. Math.­Naturv. Kl. No. 12., 1-121.

Nicholls, G.D., Graham, A.L., Williams, E. & Wood, M. 1967: Precision and accuracy in trace element analysis of geological material using solid source spark mass

spectrography. Anal. Chem. 39, 584-590. Piper, D.Z. 1974a: Rare earth elements in ferromanganese

nodules and other marine phases. Geochim. Cosmochim. Acta 38, 1007-1022.

Piper, D.Z. 1974b: Rare earth elements in the sedimentary cycle: A summary. Chemical Geo/ogy 14, 285-304.

Pisias, N.G. 1976: Late Quaternary sediment of the Panama Basin: Sedimentation rates, periodicities, and controls of carbonate and opal accumulation. Geo/. Soc. Am. Memoir 145, 375-391.

Porter, S.C., Stuiver, M. & Yang, l.C. 1977: Chronology of Hawaian glaciations. Science 195, 61�3.

Prestvik, T. & Roaldset, E. 1978: Rare earth abundances in Caledonian metavolcanics from the istand Leka, Norway. Geochem. J. 12, 89-100.

Reusch, H. 1901: Nogle bidrag til forstaaelsen av hvorledes Norges dale og fjelde er blivne til. Nor. Geo/. Unders. 32, 124-217.

Roaldset, E. 1972: Mineralogy and geochemistry of Quater­nary clays in the Numedal area, southern Norway. Nor. Geo/. Tidsskr. 52, 335-369.

Roalds et, E. 1973a: Sub-till sediments in the Numedal valley, southern Norway. Bull. Geo/. Inst. Univ. Upps. 5, 13-17.

Roaldset, E. 1973b: Rare earth elements in Quaternary clays of the Numedal area, southern Norway. Lithos 6, 349-372.

Roaldset, E. & Christie, O.H.J. 1975: Lanthanidenfrac­

tionierung in Tonen und einigen koexistierenden Mineralen. Schw. Min. Petr. 55, 191-200.

52 E. Roaldset

Robert, M. 1973: The experimental transformation of mica

toward smectite; relative importance of total charge and tetrahedral substitution. Clays, Clay Minerals 21, 167-174.

Ronov, A.B., Balashov, Yu. A. & Migdisov, A. A. 1968: Geo­chemistry of the rare earths in the sedimentary cycle. Geochem.lnt. 14, 1-17.

Rosenqvist, I. Th. 1973: Sub-moraine deposits in Numedal. Bull. Geo/. Inst. Univ. Upps. 5, 7-12.

Rueslåtten, H.G. 1976: En kvartærgeologisk kartlegging av Dagaliområdet med mineralogisk undersøkelse av

podsolforvitring i moreneavsetningene. U npubl. cand. real. thesis, Universitetet i Oslo, 181 pp.

RØnningsland, T.M. 1976: Mineralogi og geokjemi av resente

leirsedimenter i Skagerak og Kattegat. U npubl. cand. real.

the sis, Universitetet i Oslo, 173 pp.

NORSK GEOLOGISK TIDSSKRIFT l (1980)

Schmith, S.J. & Scott, A.D. 1973: Extractable potassium in Brundite illite. Soil. Sei. 116, 84-93.

Skemton, A.W. 1953: Soil mechanics in relation to geology.

Proceedings Yorkshire Geo/. Soc. 29, l, 33�2.

Taylor, S.R. 1965: Geochemical analysis by spark source mass spectography. Geochim. Cosmochim. Acta 29, 1243-1261.

Vorren, T.O. 1972: Interstadial sediments with rebedded in­

terglacial pollen from inner Sogn, West Norway. Nor.

Geo/. Tidsskr. 52, 229-240.

Vorren, T.O. & Roaldset, E. 1977: Stratigraphy and lithology

of Late Pleistocene sediments at Møsvatn, Hardangervidda, South Norway. Boreas 6, 53�9.