Benthic foraminiferal evidence of environmental change in the Skagerrak over the past six decades

Benthic foraminiferal evidence of environmental change in the Skagerrak over the past six decades

ELlSABETH ALVE

Alve,E. 1996:Bent hicforaminiferalevidenceof env ironmentalchangein the Skagerrakoverthe pastsixdecades.

Nor.geol.unders.Bull.430,85-93.

High resolution analysesof fossil bent hicforaminiferalassemblageshavebeenperformedontwoshort «40 cm) sediment coresfromthe deep Skagerrak Basin,NE North Sea.21 0Pb-datingsofreplicate coresfrom thesametwo stations havemade it possibleto calculat eaccumu lation ratesof bot hsedime ntsand bent hic for aminiferaltests, and the vertical(historical)assemb lagedist rib utio nsarecompare dwit hsedimentsurfaceassemb lagescollectedin 1937 and1992/93.Thecore datashowthatthe accumu lat ion rat e of agglutina tedtests was relat ivelyconstant from about 1770to 1940-1950andthat it hasincreasedabout5-6timessince thelate 1960'scomparedto averagepre- 1940values. Thisincreaseis duemainly to increasedproduct ion inthe 4agglut inated speciesHapl oph ragm oi des bradyi,Egger elloidesmedius,Saccam m inaspp.,and Trochamminopsispusillus,of whichthelast2 were presentonly asaccessoryspeciesbeforethe1960's.The 3 calcareousspeciesPulleniaosioensis,Nonio nellairidea,andCassidulina laeviqata,whichdomina tedthedead assemblagesprior to theearly1970's,allshowsubsequent reducedaccumu- lat ionrates.Despitepossibleerro rscaused by postmo rtemdestr uctionofsome agglutina tedcompo nents,their overallpreservatio nseemsto be good eno ugh toreflecttherealfaunal conditionsoverthelast coupleof cent uries.

The faunaldevelopment,asreflectedbythecore data,isin good agreement wit htheconclusionsbasedon the comparativest udy ofsurface assemblagescollectedin 1937and 1992/93.Thatis,the bent hicforaminifera lassemb- lages inthedeep SkagerrakBasin have changed,and according tothe coredata,these changeshave occurredpri- marily sincethe early1970's.

ElisabetbAlve,Departmentof Geology,Universityof 0510,p.a.Box7047 Blindern,N-0376 0 510,Norway

Fig.7.Bath ym etricmap(contour interval 200m)of investigationarea sho- wing thelocation andyear ofcollectionofthedeep Skagerra k Basinsurface samp lesand the positionof thetwo7993sedimentcores.

In order to evaluate what effects variou skinds of human activity (e.g.,effluent discharges and physical changes) have possibly had on the marine environment,details are needed about previous environmental conditions.

However, in most areas such information is lacking or very sparse, and the only way to gain aninsight into past conditionsis to interpretthe record which is storedin the sediments.

A study of benthic foraminiferal distribution and abun- dance from surface sediments (0-2 cm) collected by H6glund in the Skagerrak in 1937 (taxonomydiscussedin H6glund, 1947) compared with those collected in 1992/93 (Alve & Murray 1995) indicates that faunal changes have occurred, particularly in the deep SkagerrakBasin. These changes were interpreted asindi- cating environmental change. The 1937 surface sedi- ment sare now burried at a depth of 9-18cm below the modern sedimentsurface.In order to test to what extent these changes are reflected by the fossil assemblages in the sedimentary record (i.e., in cores penetrating sedi- ments deposited >50years ago), foraminiferal analyses have been carried out on two short « 40 cm), 210Pb- dated sediment cores from the deep Skagerrak Basin.

A sum mary of previous foraminiferal studies in the Skagerrak has been given by Alve&Murray (1995) and is not repeated here.

Benthic foraminiferal tests are preserved in most mari- ne sedimentsbut it is known that in certain cases, some calcareous tests are lost through dissolution (e.g., Scott 1975,Alve&Nagy 1986) and that some agglutinated taxa with organic or weakly mineralised cement are lost through early diageneticdestruction (e.g.,Douglas et al.

1980; Mackensen et al. 1993).Consequently, studies in which changes in foraminiferal assemblages are docu- mented both by down core variations and by compari-

59'00'

ss'oo

400 200 9'00'

*

^Hbglund,1937

• Alve&Murray,1992

o

Alve&Murray,1993 + Cores,1993

NORWAY

58'30'

Introduction

86 Elisa be thA/ve NGU-BUll430.1996

sons of surface samples collected today and in the past (now equ ivalent to a subsurface layer in the cores) are im portant forevaluat ing the applicabi lit yofforaminiferal analysisin st udi es oflong termenvironmentalchange.

Mate rial and Methods

Field and laboratory procedures

Thesedime ntcoreswere collectedfromtheSkagerrakat 652 m(station 56)andat 595 m(st at io n74)water depth (Fig.1)inJuly 1993(Universit y ofBergen cruise no.9307) by meansofa Niernistoegravitycorer(inner diameter50 mm). Im medi ately after collect ion, most of the water overlyingthe sedi ment in the core liner was carefullysip- honed offslowly.The sediment was then gently pushed upthrough the core liner and the lastfew millimetres of waterwascarefullyremoved wit ha pipette and transfer- red to the surfacesample(0.0-0.5 cm)container.The sur- face1cm was sectionedinto two 0.5 cmslices,andthe2- 6 cm and6-20cmintervalsweresecti on ed in 1cmand2 cm slices,respectiv ely.Finally,the restofthe2 coreswere sect ioned in 4cm (core56) and 5 cm (core 74) slices, respectively.The7samples from the upper 6cm of core 74 were gently mixedwith 70%ethanol, whereasthose from core 56 were kept at ambient temperat ure (6-8°C) until theywerewet- sieved(63 urn sieve)in ambient tem- peratureseawater within3 days of co llect io n.Fift y ind ivi- duals from eachsample were picked forATP (adenosine triphosphate) ext raction (Alve 1995)and the remaining partsofthese7sampleswere preserved in70%ethanol.

Samplesfromthe deepercore levelswerekeptrefrigera- ted on the ship and frozen immediately after return to the on-shore labo ratory. All ethanol-preserved samples wereprocessedby washing them on a63/lm sieve and stainingthe residueswit h rose Bengal for about one hour before they werewashed againon the same sieve and dried at 50OC.The frozen sam plesfrom 6-20 cm core depths were thawe dinethano l and processedas descri- bed above.Aslivin gindivi du alswerenot expected to be present below20 cm,thedeeper sampl eswerenotstai- ned but wereotherwiseprocessedin thesameway.

At least 250 dead indivi dualswere picked andidenti fi- ed from each sample.Fragmentsof tubu lar and branc- hing forms were treated as a separat ecategoryand have not been incl uded in thecalculations(fordiscussion,see Murray&Alve1994).

Dati ngsand calcula tion of foraminiferal testaccum ulation rates

Variousmethodshavebeen used to determine therate of sedi men taccumulation inmodern sedi ments(seereview by vanWeering et al. 1993).The most reliable are those based on210Pband in the present study,replicat e cores from both stations have been dated by Helmar Kunzendorf using the210Pb-met hod.Thedating wascar- ried out on1cm thick slices through out bothcores.From the result s,the calculated sediment accumu lation rates are 0.23and 0.15cm/yrforcores56 and 74,respectiv ely.

Previouslypublished ratesfor the studyarearangefrom 0.08 to 0.22 cm/yr (van Weering et al. 1987, 1993).

Consequently,the valuesfor thetwonewl yinvesti gated

Tab le1.Accum ulationrates andrelativ eabunda nceof im portantspecies(seetext).and calculatedfau nal parametersof dead asse m blagesin core56.•=

Ages aregivenas who leyea rsandreprese nt themiddleof eachdepthinte rv al.

Core 56.dead assemblages

Depth(cm) 0-0.5 0.5-1 1-2 2-3 3-4 4-5 5·6 6·8 8-10 10-12 12-14 14-16 16-18 18-20 20-24 24-2828-32 32-37 Year" 1992 1991 1988 1984 1980 1976 1973 1969 1963 1956 1949 1945 1940 193 1 1917 1902 1888 1871 No.ll 0cm²/yr

Ear/andam minasp. 6 8 17 10 13 28 16 4 7 4 9 24 5 4 12 5 2 4

E. medius 12 17 16 25 21 27 22 15 15 17 7 9 3 7 11 1 3 5

H.bradyi 71 64 44 47 58 39 38 29 19 15 15 32 20 17 24 10 6 8

Saccamminaspp. 112 58 27 40 48 28 18 2 3 2 0 2 1 0 0 0 0 0

T.pusil!us 49 27 29 32 52 28 12 13 17 1 0 4 4 2 4 2 1 4

C.laevigata 5 3 2 9 10 13 16 46 12 16 29 28 30 21 33 9 34 26

N.iridea 25 5 5 15 42 64 35 80 88 72 82 84 60 52 83 60 36 54

P. os/oencis 31 19 7 28 22 39 22 51 40 64 113 145 57 78 62 66 59 42

S.lusiformis 2 1 2 2 6 7 7 16 20 14 16 50 26 20 21 13 16 27

%va lues

Earlandammina sp. 1 3 9 4 4 7 5 1 2 1 2 4 2 1 3 2 1 2

E.medius 3 6 8 9 6 7 6 3 4 5 1 2 1 2 3 0 1 2

H. bradyi 17 23 21 16 16 9 11 6 5 4 3 6 6 6 6 4 2 3

Saccamminaspp. 26 21 13 14 13 7 5 0 1 1 0 0 0 0 0 0 0 0

T.pusil!us 12 9 14 11 14 7 4 3 5 0 0 1 1 1 1 1 0 2

C.teeviqete 1 1 1 3 3 3 5 10 3 5 6 5 9 7 9 3 13 9

N.iridea 6 2 2 5 11 16 11 17 25 21 17 16 18 18 22 22 14 19

P.ostoensis 7 7 3 9 6 9 6 11 11 19 23 27 18 27 16 24 23 15

S.fusiformis 0 0 1 1 2 2 2 4 6 4 3 9 8 7 5 5 6 10

No.counted 278 243 270 253 250 274 247 256 261 275 268 250 250 284 31 257 282 260

Testslcm³ 128 84 76 120 154 143 83 126 122 132 117 105 12 1 136 160 98 97 140

%agglutinated 74 76 82 64 61 51 52 29 26 19 17 20 22 18 19 11 10 14

Tests/l0cm²/yr 426 282 204 292 375 407 334 466 349 343 489 540 327 290 386 274 260 281

AggI.ll0cm²/yr 314 214 167 185 229 206 173 133 90 64 84 108 72 51 75 30 26 40

NGU-BULL430,1996 Elisabeth Alve 87

Table2.Accumulationrates and relative abundanceof important species(see text), and calculatedfaunal parametersofdead assemblages in core74."=

Ages aregivenas whole years and representthe middleof each depthinterval.

Core74,dead assemblages

Depth(cm) 0-0.5 0.5-1 1-2 2-3 3-4 4·5 5-6 6-8 8-10 10-12 14-16 18-20 20-25 25-30 30-35 Year" 1991 1988 1983 1977 1972 1967 1961 1952 1939 1924 1897 1869 1845 1810 1777 No.!10cm²/yr

E.medius 15 11 13 34 37 23 19 7 8 4 3 5 2 3 6

H.bradyi 40 45 29 48 28 17 13 6 4 4 14 7 6 8 7

Saccamminaspp. 45 29 9 15 3 1 0 0 0 0 0 0 0 0 1

T.pusillus 30 18 10 20 19 15 6 5 3 0 0 0 0 0 0

B.skagerrakensis 10 33 29 59 68 56 60 21 16 10 46 33 6 5 7

C./aevigata 3 2 3 7 21 19 29 34 36 27 26 18 27 23 12

N.iridea 8 4 4 29 45 44 40 42 74 60 90 27 27 27 34

P.ostoensis 21 25 19 35 63 52 54 37 78 49 68 59 39 23 45

S.fusiformis 9 9 5 8 6 8 4 11 16 9 13 10 7 9 10

%values

E.medius 6 5 8 10 10 7 6 3 2 2 1 2 1 2 3

H.bradyi 15 18 18 14 7 5 4 3 1 2 4 3 3 5 4

Saccammina spp. 17 12 6 4 1 0 0 0 0 0 0 0 0 0 0

T.pusi!lus 11 7 6 6 5 4 2 2 ¹ 0 0 0 0 0 0

B.skagerrakensis 4 14 18 17 17 16 20 9 5 4 13 13 3 3 4

C./aevigata 1 1 2 2 5 5 10 15 11 11 8 7 15 15 7

N.iridea 3 2 2 8 11 13 13 18 22 24 27 11 15 17 20

P.os/oensis 8 10 12 10 16 15 18 16 23 20 20 24 22 15 26

S.tusitormis 3 4 3 2 2 2 1 5 5 3 4 4 4 6 6

No.counted 247 234 252 248 263 255 257 252 265 288 260 317 269 277 258

Tests/cm- 179 164 106 177 192 197 189 137 254 171 246 167 126 106 120

%agglutinated 65 59 54 45 33 28 17 17 7 8 10 8 16 15 15

Tests/jOcrne/yr 267 244 163 348 392 340 300 231 344 250 340 248 180 157 173

Agg1.l10cm^2lyr 174 143 88 157 130 93 51 39 25 19 34 20 29 24 26

cores are in good agreement with these previous fin- Accu m ulati o nrates (no./l Ocm

2/

y r) dings.The valuesreflect the high varia bilityin the sedi- 0 100 200 300 ment accumulation rates in the area and, as a consequen-

1990

.:':~" .".'.~~:',~

^{..............•}

ce, core 56 penetratedsediments deposited since about

1870,whereas core 74 penetrated back to about 1770 1970 _.~....^"

""

even though the cores we re of comparablele ngt h.

The accumulation rates of dead,benthic foraminifera 1950

. . ^:~:::.

^{Core 56}

(defined as the number of testsper 10 ern?per year)were

calculated as follows.Foreach sample,the totalnumber 1930 ^41. of dead foraminiferal tests was divided by the core area

•

(in crn-),multiplied by 10,and finally divided by the num- 1910 ber of ye arsofse d ime n ta t ion represented by tha t sam - .... ₁₈₉₀ pie.This gave the number of dead tests added to the ^C':Q) sediment per 10 ern?peryear.The accumulation ratesof

>-

₁₈₇₀

individual species and of dead agglutinated test s were 70

calculated in thesa me way. 1850 60

'"

⁵⁰

1830

-,

^{§ 40}

Re sults

^{Core74 "230}

1810 z 20

The faunaldata,re levan t for the present discuss ion,are 10 1790

presented in Tables 1 and 2. According to the 21OPb- 0

1937 1992/93 dating, the actual 1937 sea-bed su rface now lies at a 1770

depth of 17-18cm (co re 56)and 9-10cm(core 74)below

Fig.2.Accu m ulatio n ratesof dead agglutinated testsincores56and74.

the curre nt sea floor. In sedime nts deposited befo re Histograms show theaverag econcentrationof agglutinatedtestsin deep

about 1910 (Fig. 2),that is below 24 cm in core 56 and Skagerrak Basin surfacesamp les collected in 1937and 1992/93(surface

below14cm in core74,the accumulation rat es of agglu- data from Alve&Murray1995).

tinated testsare fairlyconsta nt(20-40test s/10cm-/yr).At levels correspond ing to the late 1960's,the rates have

accu mul ation rates of 314 (co re 56)and 174 (co re 74) increased to about 100-1 30 tests/10 ems/yr .Su bsu rface

loca l minima are succeed ed,in both cores, by maximum agglutinated tests/l0 cm2/yrin the mostrece nt ly deposi- ted 0.5cm of sediment.

88 ElisabethAlve NGU-BUl l 430,1996

% %

0 10 20 30 0 10 20 30

1990

.-

-

Core 74 Core 56

1970

,.

1950 ?

" -1

1930

~

1910 ^/

,,' /

@1890 ^/

- : :

v

,

>-< 1870 _{.: "} ^\

1850 ^{- +--} H.bradyi

~ - Saccamminaspp

1830 ...•.. _N.iridea

1810 ^--+- Piosloensis

1790 1770

Fig.3.Relativ eab u n d an c e of the4 mostfrequently occu rrin gspecies th ro u gh out cores56 and74.

Nine speciesmake up_~5%of thetotaldeadassembla- gein at least two sam pleswit hin eachcore.Eigh tofthese are the sameinboth coresand the relati veabundances of the 4 most common species(i.e.,domi nantin at least one sample), except Brizalina skagerrakensis (Qvale &

Nigam),areshow nin Fig. 3.Thereasonfor omitting this speciesfromthediagram is discussedbelow.The deeper parts ofthe cores (>4 cmin core56; >6cmin core 74)are dominated by one or the other of Nonionella iridea Heron-Ali en & Earland and Pullenia osloensis Feyling - Hanssen. In core 56, Haplophragmoides bradyi (Robert son)dominatesthe core interval 0.5-4.0cm with frequentSaccam m inaspp.and Trochamminopsispusillus (Hoqlund), whereasSaccamm ina spp.dominate the sur- face0.0-0.5cm with the other two as frequent compo- nents. In core 74,B.skagerrakensis dominatesthe core interval2-6 cm wit h frequentP.osioens is,N.iridea,H.bra - dyi ,andEggerelloi des medius(Hbglund). The surface2cm are generally dominated by H. bradyi with frequent Saccammina spp., B. skagerrakensis, P. osloensis, and T.

pusillus.

The comparative study of surface assemblages from 1937 and 1992/93 (Alve & Murray1995)indicatedthat the greatest changeshad occurred in 4 agglutinated and 3 calcareous taxa. Thecalculat ed accumulati onratesof the 4 agglut inated taxaSaccam m in aspp.,T.pusillu s,E.medi- us,andH.bradyi,arerelativelylow in both coresin sedi- ments deposited before about 1950 with ::;1,::;5, 1-11, and generally 4-20 tests/10 crn-/yr,respectively (Fig. 4).

While E. mediusshows maximum values (27-37tests/10

cm-/yr)in bothcores atcore depthscorrespo nding tothe 1970's,theother 3 generally show maximum valuesin the upper 1-2cm:Saccamminaspp.45-112;T.pusillus 30- 49;H. bradyi45-71 tests/10 cm-/yr .

Of the 3 calcareous taxa, Cassidulina laeviga ta d'Orbigny generally has accum ulation rat es betwee n 15 and 35tests/10 crn-/yrin pre 1970core lev el s,whereas thevalues forP.osloensisandN.irideagenerallyliein the range 30 to90tests/10 cm-/yr.Min imumvalues for all 3 species occur in the upper 1-2 cm:C.laevigata 2;P. oslo- ensis7-19;N.iridea4-5tests/10cm²/yr.

Discussio n

Accumula tion ratesof agglutinatedtests

Theaccumula ti onrates ofagg lut inated testsindicaterea- sonably stabl e production from the late 1700's until 1950-1960,although a mino r increaseisindicated in core 56 from about 1910(Fig.2).However,both coresshow a pronounced increase from the late 1960's; by the late 1980's to early1990's(represented by the accum ulati on ratein thesurface0-2cmin core56 and 0-1cmin 74),the rate hadincreasedby factors of4.7(core56)and 6.3(core 74),compared toaverage pre-1940 values.These values arein good agr eementwiththefindings of Alve & Murray (1995).Their com parativ estudyof benthicforaminiferal surface (0-2 cm) assemblages collected in the deep 5kagerrakBasin in 1937andin1992/93indicat ed that the

NGU-BULL430, 1996 E/isabeth A/ve 89

Acc umulation rates (no.Zl Ocm-yyr)

o

^{10 20 30 40 50 0}

^Acc umulatio n rates (no./ l Oc m 2 /y r)

10 20 30 40 50 60 70 80 90

112

1950 1970 199 0

1930

I 1910 I I

~

~ 1890 \

~ \

0

Core 56

>--

^\

1870

r

\ ¹⁴

I

•

¹⁹³⁷

I 12

I£l

^1992/93

1850 I

T

10

I

1830 I ^{r .}

Core 74

^{,....c 8}

\ ^o

--

I 0 6

1810

•

_\ ^{- +--}

^{E . medius z}

₄

\

--- ^{H. bradyi}

1790 \

Sa ccammina spp

²

\

...*...

T. pusillus

⁰

----+-

1770 H.bradvi Saccam. T.pusillus E.medius

Fig.4.Accumulation rates of 4key agglutinat ed species incores56and74.Histograms showtheir average concentrationin deepSkagerrakBasinsurface samplescollected in 1937and1992/93 (surfacedatafromAlve&Murray1995).

average abundance of agglutinated tests hadincreased by a factor of 3.8 during that timeperiod.As thiscompari- son was based on only 3 samples from 1937 and 12 sam- ples from 1992/93 the factor of 3.8should be regarded merely as an approximation.On the other hand, it can be argued that some of the agglutinated tests have been lost through early diagenetic destruction so that the agglutinatedtest concentration is somewhat reduced at deepercorelevelscompared to the time when they were deposited.

Aggluti nated tests whichare weakly held together by organic materialdo not survive long after death (Murray 1991).Douglas et al.(1980) suggested that the organic matter probablyisoxidised within the surface layerofthe sediments,andSid ner&McKee(1976)indicated that the vertical distribution of iron-richagglutinated taxa is con-

trolled by geochemical factors rather than ecological ones. Additionally, some agglutinatedspeciesdonotsur- vive sample processing simplybecause their tests are too fragile.Undoubtedly, some of the more fragile speciesin the deep Skagerrak Basin either rapidly disintegrated after death due to oxidation or bacterial attack of the organic material, or were destroyed during processing.

Alve & Murray (1995) noted that for instance Haplophragmoides membrana ceum Hoqlund,which was common in the stained assemblagesin the deep basin, was rare in thecompanion dead assemblages, probably dueto postm ortem dest ruct ion. However, becausethis destr ucti onseemed to happen soquickly, and because the stainedassemb lagesgenerallymade up<15%of the totalassemb lages(staine d

+

dead in upper0-2cm)upon which the comparison between surface samples was

90 Elisabeth A/ve NGU-BULL 430,1996

Accumul ation rates (no.!10cm 2/y r) Accumu lation rates (no.!10cm2/yr)

0 20 40 60 80 0 20 40 60 80 100

1990 * ..*

'*

^~~^{l: ::: .}

... : '*

... - ^{' ... .. ... .... .. :."*}

^{" *-........}

^*

1970 -,

_<, ~. ^{.... -*-,~}_ _^:~.- -

^{. ...}

- -

^...

0°

^{. *}

--... ,

^{-* ~ -}

^{"' : ...}

... ...

^"

^.

1950

^{• *-}

\

^'..

1 145

•

^/

•

/ ¥

1930

_/

_-,

f -:

'\

I ,...

1910 I ,.... ,....

I ... ,....

•

^{<, <,}

/

^<,

l-.

1890

^<, _*-

ro /

Cl) /

Core 56

>-- ^/

/ /

1870 ,

^"#

\

14

\

• ^{1 937}

1850 _\ 1 2

Eill ^{1992 /93}

t.

I ¹⁰

1830 I I . :

^{( " j}

--

^c:u

8

I: - -

^C.laevigata

--

⁰

⁶

1 810

^{~. Z}

/ --

^{N. iridea}

⁴

/

_...*... P. osloensis

/ 2

1790

/

/

Core 74 ⁰

• ^'*

Pullcnia spp . N. iridc a

C. laevigata

1770

Fig.5.Accumulationrates of 3key calcareousspeciesincores56 and 74.Histogra msshowtheir averageconcentrationin deepSkagerrakBasin surface sam- ples collected in 1937 and1992/93(surface data fromAlve& Murray1995).

based,discrepancies caused by the presence of fragile species were almostelim inated.In thedown core compa- risonsmadehere,only the deadassemblages wereconsi- dered.Consequ ent ly,themostfragile species had alrea- dybeen eliminated fromthe surfa ce sediment assembla- ge data,before the down corecomparisons were under- taken.

Loss of agglutinated tests in the surface sediments have also been report ed from the East Pacific Rise

(Douglas& Woodruff1981),thesout hern Californiabor- derland (Douglas et al. 1980), the northwest Atlant ic Ocean (Schroder 1988), the eastern Weddell Sea (Mackensen&Dougl as1989,Mackensenet al.1990),and beneath theSouthAtlant icPolar Fron t (Mackensen et al.

1993). Schroder (1988) distinguished 3 categories of deep-sea agg lutinated speciesaccording to their preser- vati on potentialand Mackensen et al.(1990, 1993)went asfar as exclud ing fromtheir«potential fossilassemb le-

NGU-BULL430,1996

ges» all agglutinated species that they considered as non-resistant.A feature common to the samples analysed in most of theseinvest igati ons is that they were collected from relatively deep-waterenvironments (>1000 m).On the other hand, down coreinvest igati ons in more margi- nal marine areas have reported abundant agglutinated assemblages well below the surface few centimetres(i.e., throughout140-170 cm cores in Drammensfjord, Norway (Alve 1991), throughout cores of up to >240 cm in Saguenay Fiord,Canada(Schafer et al. 1991),and down to 7 m below the sediment surface in Lake Melville, Labrador(Vilks&Mudie 1983).Denne&Sen Gupta(1989) concluded that«t aphonomic processes are least activein areas of rapid sedimentation, where the bioturbated zone is thin, and under oxygen-deficient conditions».

Consequently, it is probable that the more selective pre- servation of agglutinated tests in the deep-sea compared with that in marginal marine environments is, to a great extent, related to their differing sedimentation rates.

Another possibility is that many typical shelf and margi- nal marine agglutinated species may have a higher pre- servation potentialthan that of deep-seataxa, as a gene- ral adaption to more stressed environments.

The deep Skagerrak Basin has fairly rapid sedimentati- on rates (>1 mm/yr) and sediment porewater oxygen profiles in sediment cores from 27 stations in the Skagerrak, including the deep basin, showed that the oxygenated zoneis thin(3-20 mm,Bakker & Helder 1993).

Consequently, if a rapid sedimentation rate and a thin oxygenated zone are regarded as important criteria for the preservation of agglutinatedtests,the agglutinated tests in the deep basin sediments should stand a good chance of not getting lost. Additionally, the deep Skagerrak Basin can be geomorphologically considered as a silled fjord (separat ed from the rest of the North Sea by a sill at about 270 m water depth), and most of the common species are also frequent components in the adjacent fjords (i.e.,Ovale et al. 1984) where abundant and diverseagglutinated assemblages are preserved in the sediments (Alve 1991, unpubl. data).

To summarise, it can be concluded from the above- mentioned factors,and the fact that the core data show relatively constant accumulation rates of agglutinated tests for nearly 200 years before the recent increase, that the core data reflect the originalfaunal development rea- sonably well.Furthermore,the down core data support the conclusion drawn from the comparative study of sur- face assemblages of 1937 and 1992/93thatthe abundan- ce of agglutinated foraminifera has increasedin the deep SkagerrakBasin duringthelast sixdecades.

Changesin the faunal composition

The comparison with Hoqlund's1937 surface assemblage

data indicated that apronounced inc re ase ha soccurred

in 4 agglutinated species in the deep basin (see lower right-hand diagram in Fig. 4 and Alve &Murray 1995).

Two of these, Saccammina spp. and Trochamminopsis

Elisabeth Alve 91

pusillus, were not recorded by Hoqlund in this area.

However, scattered occurrences of both at pre-1940 core depths, show that they have been present there for a longer timeperiod,but only as accessory species(Fig. 4).

The two other species, Haplophragmoides bradyi and Eggerelloides scabrus,were the most frequently occurring agglutinated species in Hoqlund's deep basin samples.

This is also in good agreement with the core data,as they are the most common agglutinated taxa at levels corres- ponding to sedimentsdeposited before about 1960.

The overall most abundant taxa in Hoqlund's deep basin 1937 assemblages were Cassidulina laeviqata, Pulleniaspp. andNonionella iridea(as hisNonion labrado- ricum?).Accordingly,the same species are also the most abundant in both cores in sediments deposited before about 1970.However, the surface comparison indicated an increasing trendin the abundance ofPulleniaspp. and N. irideaand a minor decrease inC.laevigata(lower right- hand part of Fig. 5), whereas the core data show reduced frequenciesin all (Fig. 5).This discrepancy is probably due to the fact that Hoqlund analysed the samples in a wet state,and probably overlooked a number ofPulleniaspp.

(mainlyP.osloensis)and N. iridea specimens,since they are small (typically< 190 urn greatest diameter;Feyling- Hanssen 1964) and thin-shelled and relatively transpa- rent in water. Consequently, it seems that the core data givemore reliable information about the faunal changes than the comparison with Hoqlund's surface assembla- ges. On the other hand,C.laevigata,whichis a bigger and thicker shelled species,shows the same decreasing trend towards today'sconditions in both the comparative sur- face and core studies.

Alve & Murray(1995) suggested that active carbonate dissolution is taking place in the deep Skagerrak Basin today (presence of corroded tests),and the recent decre- asing trend in the accumulation rates of calcareous tests in the two cores analysed seems to reinforce this observa- tion. Thistopic will be further focused upon in a later paper following completion of more core analyses.

When considering the historical development of the relative abundances of the most common taxa, the core data show that,except for a dominance ofBrizalina ska- gerrakensis between the late 1950's and 1980 in core 74 (Table 2),a pronounced faunal shift from P.osloensisand N. iridea dominated assemblages to H. bradyi and Saccammina spp. dominated assemblages occurred during the 1970's(Fig. 3;Tables 1 and 2).The subsurface dominanceofB. skagerrakens is,which is only an accesso- ry species in core 56, brings up the topic of temporal mixing of faunas due to differences in vertical habitat preferences. It is now generally accepted that certain infaunal benthic foraminiferal species may live at various depths below the current sediment surface (e.g., Corliss 1985,Gooday 1986, Mackensen &Douglas1989, Alve&

Bernhard 19 95).Deepin fa u na ltaxalive in sedimen ts la id

down decades ago.Thus, as also pointed out by Murray (1995), the final dead assemblages may consist of con- temporary deep infaunal tests plus epifaunal and shallow

92 EJisobeth Alve

infaunaltests which died decadesago.In core 74,abun- dant and nearly monos pecifi c living B. skagerrakensis assemblegswere present at 8-20cm core depths(Alve 1995). In contra st to this, B. skagerrakensis had its live maximum abundance in the upper 0.5 cm of the sedi- mentsin acore fro m thesamearea(621mwater dept h) investigated by Corliss &van Weering (1993).Based on thesefinding s,itisreasonableto assumethatthe subsur- facedominanceofdeadB.skagerrak ensisin core74does not reflectthat it actuallydominatedthe original assemb- lages deposited between the late 1950's and 1980.It s dominance isprobably rath eraneffect of its temporally changing vertic al positionin thesediments causing it to leave itsempty tests together with non-contemporary deposit ed assem blages. Exclusio n of B. skag erra kensis from the deadassemblag edataduringcalculat ions ofthe relative abundances did not alter the signals from the otherdomin ant species concern ingthe tim ingof the fau- nal shift. How ever, environmental misinterpretations of thecore interval whereit dominateswereavoided,due to theknowledge abou t its extr em ely chang ing vertical living occurrences in thesediments inthis area.

Sig nificanceofthe fau na /changes

The faunal changes indicated by the com parative study of Alve &Murray (1995) based on surface assemblages collected in 1937 and in 1992/93 have now been confir- med bythedow ncore investig ati ons.The questionthen arisesasto what these chang es actua lly mean? Thereis limitedinformation availableconcerningthe reasonswhy agglutinated assemblage s suddenly should increase in density.How ever,it is worthwhi le ment ion ing that core stu di es of sediments deposit ed over the last 400-500 years in Frierfjo rd near Porsgrunn (Fig. 1), southern Norway,showe dan increased abund ance of agglutina- ted taxaduringthe'saw-m ill period' (1600,1700 and init i- al parts of the1800's),befo re they disappeareddue to the develop ment of nearly permanent ly anoxic conditions (A lve ,unp ub l.data).Furthermo re,itisnoteworthythat,in thesamewayas in theSkagerrak,T.pusil /us,andto some degreeH.bra dyi,were amon g the agg lut inated species which showed increased abundance, whereas at the sametimeC./aevigata,whichpreviouslyhad dominated the assembl ages,showed acleardecrease.In Frierfjord, thisfaunal shift is interpreted to reflect poorlyoxygen a- ted condit ion s in thesurfacesedimen ts due to the comb- inedeff ect sofinfrequentdeep-wat er renewalsandincre- asedload of organic material.

Drasticchangesin the foramin iferalassemblageshave alsorecent ly been note d atcore levelsdated to theearly 1970'sin the southern Kattegat,bet wee ntheSkagerrak and theBalti c(Christiansenetal.1994).It wasspeculated that these changes were connected to documented wind- indu ced highe r salinities and declining bottom wateroxygen concentrations.The timing ofthe southern Kattegat faunal change s fits surprising ly well with the timing of the changes in the deep Skagerrak Basin.

GU-BULL430.1996

However,it would befar toospeculative to suggestthat these changesarelinked to some overall environmental change s that would affect two such different,alth ou gh nottoodistant,areas.

The deep Skagerrak Basin isthe main deposito ry of fine-gr ained particulate material in the North Sea (e.g., van Weeringet al.1987).Addit io nally,becauseofitsphy- sical properties(sill at about 270 m water depth) and general watercirculationpatternswhich prevent efficient deepwate rexchange(on average,deepwater renewa lis every 25th month;Aure& Dahl 1994),the bot to m envi- ronm entisprobablymore sensit iveto chang ing fluxesin nutrientsandorgan icmaterialthanwell-fl ushed areas.It isnot clearto what extent the yearlyinfluxof anthropo- genically induced nutrients and organic matter to the North Seahaschanged overthe past50-60 yearsbut con- side ring the generalindustrial and urbandevelo pm ent in northern Europe, it isreasonableto assume that it has increased.

At the present stage, it is difficult to draw any firm conclusionsabout what kind of environme nt al changes might have caused the observed faunalchangesin the deep SkagerrakBasin.However,the fact that the forami- niferal assembla ges have changed, especially over the last 20-30 years, isan int eresting observation in itself.

Further investi gati ons are needed in the area to try to explain what the reasonsmightbe.

Conclusions

Detailed foraminiferal analysesof two 21OPb-dated cores fro m the deep Skagerrak Basin which penet rate sedi- mentsdeposited sinceabout 1770and 1870,respective- ly,have shownthat :

(1) The accumulationrate ofagg lut inat ed foraminiferal tests was relative lystab leuptoabout 1950-1960 andhas increased about 5-6time s sincethelate1960'scompared toaverage pre-1940 values.

(2) The in c re ased accumulatio n rate of ag g lutin a ted tests isprimarily due to increasesinHap/ophragmo ides btadyi, Saccam m ina spp.,Trocham minopsispusillus,and Eggerello ides medius.Beforeabout 1950,H.bradyi andE.

medius were the most common agglutinated taxa,whe- reas Saccam mina spp. and Tro chamminopsis pusillus occurred onlyasaccessory species.

(3)A drast icfaunalchang efrom assemblages domina- tedbythe calcareousPull eniaosloensis,Noni onellairid ea andCassidulina /aeviga tato theabove-me ntionedagg lu- tinated speciesoccurred during the early 1970's.Since then,reducedaccumulatio n rateshave been recor dedfor all3species.

The generaltrendsconcerningfaunaldevelopment in the deep basinsupportthe conclusion sdrawn from the comparative st udy of surface assemblages collected in 1937and 1992/93(Alve&Murray1995).The discrepanci- esconcerningthetwo calcareo ustaxaP.os/oensisandN.

iridea are probably due to an underestimatio n of their

NGU-BULL430, 1996 ElisabethAlve 93

presence in the 1937 sam ples, Consequently, the core data seem, in some respect s, to give a more reliable impression ofthe real faunal changes than the surface samplecomparison.

This is the first st udy where long-term changesin fora-

miniferal assemblag eshave been documented both by

down core variationsand by comparisonsof surface sam- plescollected today and >50 yearsago.Thesefindings have important impli cati ons forfuture invest igationsas they demonstratethe applicabilityofforaminiferal analy- ses in studiesof long-te rmenvironmentalchange.

Acknowledg ements

This workwascarriedout under cont ractto the Geologi cal Surveyof Norway (NGU)and theInsti t ute of MarineResearchin Bergen.Iwould like tothank NGUand theUniversit y of Bergenfor the oppo rt unity to participat e in the1993cruiseof the'HakonMosby',Helmar Kunzendo rf for kind ly placing his 210Pb dating-results at my disposal, the Department of Geology,University ofOslo,fortheuse ofitsfacilities, and Bruce Corliss,KarenLuise Knudsenand John Murray for helpf ul commentson the manuscript.

References

Alve,E.1991:Foraminifera,climaticchangeand pollution:Astudy of Late Holocene sediments in Dramm ensfjord, SE Norway. The Holocene1, 243-261.

Alve,E.1995:Miljostratigrafiskeundersokelser i den nord vestredelav Skagerrak over de siste 200 ar belyst ved foraminiferin nholdet i utvalgtekj erner. Nor.geol. unders.Rapportno.95.075,44pp.

Alve,E.&Bernhard,J.M.1995:Vert icalmigratory response of benthic foraminiferato cont rolled oxygenconcentrationsinanexperimen- tal mesocosm.Mar.Ecol.Prog.5er.116,137-151.

Alve,E.&Murray,JW.1995:Benthicforaminiferaldistributi onandabun- dancechanges inSkagerraksurface sediments:193 7(Hoqlund)and 199 2/93datacompared. Mar.Micropaleontol.25,269-288.

Alve, E. & Nagy, J. 1986: Estuarine foraminiferal distribution in Sandebukt a,abranch ofthe OsloFjord.J.Foram.Res.16,261-284.

Aure,J.&Dahl,E.1994:Oxygen,nutrients, carbon and water exchange in the SkagerrakBasin.Continental Shelf Res.14,965-977.

Bakker,J.F.&Helder W.1993:Skagerrak(nort heast ern NorthSea)oxy- genmicroprofilesandporewater chemist ryinsediments.Mar.Geol.

111,299-321.

Christ iansen, C, Kunzendorf, H., Laima, M.J.C, Lund-Hansen, L.C &

Pedersen,A.M.1994:Recentchanges in environmentalcondit ions inthesout hern Kattegat.5th MarskatMeeting,Abstracts,p.4.

Corliss, B.H.1985:Microhabitatsof bent hicforaminifera wit hin deep-sea sediments.Nature314,435-4 38.

Corliss,B.H.&VanWeering,T.C E.1993:Livin g(stained) benth icforami- nifera withinsurficialsedi mentsof the Skagerrak.Mar.Geol.171, 323-3 35.

Denne,RA&Sen Gupta,B.K.1989:EffectsofTaphon om y and Habitat on the Record of Benthic Foraminifera in Mod ern Sed iments.

Palaios4,414-423.

Douglas,R.G.,Liestma n,J.,Walch,C,Blake,G. & Cotto n,M.L.1980:The transitionfrom liveto sedimentassemblagesinbenth icforami nife-

rafrom thesouthern Californiaborderland.InField,M.,Bouma,A., Coulbo rn,I.,Douglas,R.G.&Ing le,J. (eds): Quaterna ryDepositi onal Environment of the Pacific Coast. Pacific Coast Paleogeography Symposium,vol.4,Pac.Sect.,SEPM.,256-28 0.

Douglas,R.G.&Woodruff,F.1981:Deep sea benthic foraminifera.In Emiliani, C (ed.):TheSea,vol.7,TheOceanic Litnosphere,Wiley- Interscience,NewYork,1233-1326.

Feyli ng-Hanssen,RW.1964.Foraminifera in Late Quaternary deposits from the Oslofjordarea.Nor . geol.unders.225, 383pp.

Gooday, A.J. 1986: Meiofaunal foraminiferan s from the bathyal Porcupine Seab ight (northeast Atlantic):sizest ruct ure, standing stock, taxonomiccomposition,species diversity and verticaldistri- butionin thesediments.Deep-SeaRes.33,1345-13 73

Hoqlund,H,1947:Foraminiferain theGullmarfjord andthe Skagerrak.

Zool.BidragUppsala,26,328pp.

Mackensen,A. &Douglas,R.G.1989:Down-coredistributionof liveand dead deep-wat er bent hic foraminifera in box cores from the WeddellSeaandCalifornia continental borderland.Deep-SeaRes.

36,879-900.

Mackensen,A.,Futterer,D.K.,Grobe,H.&Schmiedl,G.1993:Benthic for- aminiferalassemb lages from the eastern SouthPolarFront region between35°and57"5:distrib ution,ecologyand fossilizat ionpoten- tial.Mar.Micropal.22,33-69.

Mackensen,A.,Grobe,H., Kuhn,G.&Futterer,D.K.1990:Benthicforami- niferalassemblagesfromthe easternWeddell Seabetween68and 73"5:dist ribution,ecologyand fossilizationpot ential.Mar.Micropal.

76,241-283.

Murray,J.W.1991:Ecology andpalaeoecol ogy of benthicforamin ifera.

397pp.Longman.

Murray,J.W.1995.Microfossilind icatorsofoceanwatermasses,circula- tionand climate.In:Bosence,D.W.J.&Allison,P.A. (eds),Marine PalaeoenvironmentalAnalysisfromFossils,Geol.Soc. Spec.Publ.83, 245-26 4.

Murray,J.W.&Alve,E.1994:High diversityagglutinatedforaminiferal assemblages from the NE Atlant ic: dissolution experiments.

Cushman Found.Spec. Publ.32,33-51.

Qvale,G.,Markussen,B.&Thiede, J.1984.Benthicforaminifersin fjords:

respo nse to water masses.Nor.Geo/.Tidsskr.66,325-332.

Schafer,CT.,Collins,E.5.&Smit h,J.N.1991:RelationshipofForaminifera and thecamoebian distributions to sediments contaminated by pulp mill effluent: Saguenay Fiord, Quebec, Canada. Mar.

Micropaleontol.17,255 -283.

Schroder,CJ.1988:Subsurfacepreservat ionof agglutinatedforaminife- rain the northwest Atlantic Ocean.Proceeding s of the second workshop on agglut inated foraminifera. Abhandlungen der Geologischen Bundesanstalt41,325-336 .

Scott,D.B.1975:Quantitativestudiesof marshforam iniferalpatternsin sout hern CaliforniaandtheirapplicationtoHolocenestratigraphic probl ems.Marit imeSedime nts Spec.Publ.7,153-170.

Sidner, B.R.&McKee,T.T.1976:Geochemi cal cont rols onvertical distri- but ion of iron-rich agglut inated foraminifersin Late Quaternary continentalslopesediments fromnorthwe stGulfof Mexico.Amer.

Assoc.Petrol.Geol.,Bull.60,p.722.

VanWeering,T.CE.,Berger,GW.&Kalf,J.1987:Recent sedi mentaccu- mulationin the Skagerrak,nort heast ern NorthSea.Netherlan dsJ.of SeaRes.21,177-189.

Van Weering,T.CE.,Berger,GW.&Okkels,E.1993:Sedi ment transport, resuspension and 210Pb accumulation ratesin thenortheastern Skagerrak.Mar. Geol.771,269-285.

Vilks,G.&Mudie,P.J.1983:Eviden ceforpostglacialpaleoceanograph ic and paleoclimat echangesin Lake Melville,Labrador,Canada.Arct ic and Alp ineResearch75,307-320.

Manuscript receivedMarch 7995.'revisedversion accep ted Ja nua ry 1996.