Climate Qualitative and quantitative reconstructions of surface … · 2020. 6. 19. · culation of the topographically steered North Atlantic Cur-rent (NAC) and the Norwegian Coastal

Clim Past 10 305ndash323 2014wwwclim-pastnet103052014doi105194cp-10-305-2014copy Author(s) 2014 CC Attribution 30 License

Climate of the Past

Open A

ccess

Qualitative and quantitative reconstructions of surface watercharacteristics and recent hydrographical changes in theTrondheimsfjord central Norway

G Milzer 1 J Giraudeau1 S Schmidt1 F Eynaud1 and J Faust2

1Environnements et Paleacuteoenvironnements Oceacuteaniques et Continentaux UMR5805 CNRS ndash Universiteacute Bordeaux 1Avenue des Faculteacutes 33405 Talence France2Geological Survey of Norway (NGU) Trondheim Norway

Correspondence toG Milzer (gcmilzergmxde)

Received 30 July 2013 ndash Published in Clim Past Discuss 13 August 2013Revised 4 December 2013 ndash Accepted 3 January 2014 ndash Published 12 February 2014

Abstract In the present study we investigated dinocyst as-semblages in the Trondheimsfjord over the last 25 to 50 yrfrom three well-dated multi-cores (210Pb and 137Cs) re-trieved along the fjord axis The downcore distribution ofthe dinocysts is discussed in view of changes in key hydro-graphic parameters of the surface waters (sea-surface tem-peratures (SSTs) sea-surface salinities (SSSs) and river dis-charges) monitored in the fjord We examine the impact ofthe North Atlantic Oscillation pattern and of waste water sup-ply from the local industry and agriculture on the fjord eco-logical state and thus dinocyst species diversity Our resultsshow that dinocyst production and diversity in the fjord isnot evidently affected by human-induced eutrophication In-stead the assemblages appear to be mainly controlled by theNAO-related changes in nutrient availability and the physico-chemical characteristics of the surface mixed layer Still dis-charges of major rivers have been modulated since 1985 bythe implementation of hydropower plants which certainly in-fluences the amounts of nutrients supplied to the fjord Theimpact however is variable according to the local geograph-ical setting and barely differentiated from natural changes inriver run off

We ultimately test the use of the modern analogue tech-nique (MAT) for the reconstruction of winter and summerSSTs and SSSs and annual primary productivity (PP) in thisparticular fjord setting The reconstructed data are comparedwith time series of summer and winter SSTs and SSSs mea-sured at 10 m water depth as well as with mean annualPPs along the Norwegian coast and in Scandinavian fjords

The reconstructions are generally in good agreement withthe instrumental measurements and observations from otherfjords Major deviations can be attributed to peculiarities inthe assemblages linked to the particular fjord setting and therelated hydrological structure

1 Introduction

The climate of western Norway is closely coupled to the cir-culation of the topographically steered North Atlantic Cur-rent (NAC) and the Norwegian Coastal Current (NCC) flow-ing northward over the adjoining Norwegian shelf and conti-nental margin (Loeng and Drinkwater 2007) The lateral dis-tribution and physico-chemical characteristics of the NCCwhich is supplied by brackish water from the Baltic Searivers and fjords along the Norwegian coast are linked to theprevailing wind patterns in the North Atlantic realm (Saeligtre1999 2007) The Trondheimsfjord in western central Nor-way adjoins the Norwegian Sea The fjord hydrology is sim-ilarly coupled to the NAC and NCC circulations as wellas to local riverine input from the hinterland (Sakshaug andMyklestad 1973 Jacobson 1983) Fjords are relatively shel-tered environments where biogenic and lithogenic sedimentsaccumulate at very high rates (Howe et al 2010) The fjordsediments and incorporated fossil microorganisms thereforeprovide highly resolved information on temporal changes inthe fjord hydrology related to the proximal ocean current

Published by Copernicus Publications on behalf of the European Geosciences Union

306 G Milzer et al Dinocysts in the Trondheimsfjord during the past 25ndash50 yr

variability and on the interaction of marine and continentalclimate conditions

The distribution of organic-walled dinoflagellate cysts (ordinocysts) the zygotic products of dinoflagellates in sed-iments is controlled by the surface water characteristicsmaking them widely used as paleoceanographic proxies forthe reconstruction of sea-surface parameters (eg Harland1983 de Vernal et al 1993 Rochon et al 1999 Marretand Zonneveld 2003 Zonneveld et al 2013) Dinocystsare composed of highly resistant refractory organic matterand are therefore usually well preserved in sediments (egRochon et al 1999 de Vernal et al 2000 Zonneveld etal 2007) Dinoflagellates in coastal environments are gen-erally adapted to large sea-surface temperature (SST) andsalinity (SSS) ranges due to the seasonally varying relativeinfluence of marinecoastal and continental waters In thesesettings their occurrence is therefore rather related to otherspecies-discriminating environmental parameters than in theopen ocean such as lateral and vertical salinity gradientsnutrient availability and the species-specific ability to adaptto rapidly changing hydrological conditions (eg Smaydaand Reynolds 2003 Harland et al 2004 2006 Radi et al2007 Pospelova et al 2010) These hydrological changesmay however be biased by local and regional anthropogeni-cally induced changes in the coastal and fjord environments(Saeligtre et al 1997 Dale 2001) Recent studies on surfacesediments from the Trondheimsfjord which encompass thepast 4ndash20 yr show large spatial variability in the compositionof cyst assemblages linked to the complex fjord hydrographyand topography (Milzer et al 2013) Untreated waste waterfrom local and regional industry and agriculture was directlyled into the Trondheimsfjord until 1980 The assemblageshowever do not reveal any indications of human-induced eu-trophication suggesting that recently imposed environmen-tal regulations and restrictions to reduce these nutrient inputswere highly effective (Tangen and Arff 2003) Alternativelythe exchange of Trondheimsfjord waters with the open seastrongly dilutes excess nutrient loads and the regular supplyof oxygen-rich Atlantic waters prevents hypoxic or anoxicconditions at the bottom (Tangen and Arff 2003)

In the following study we investigate recent changes inthe dinocyst assemblages in the Trondheimsfjord over thelast 25 to 50 yr from three sediment cores located along thefjord axis We discuss the downcore distribution of dinocystsin view of naturally and anthropogenically induced changeswith regard to SST and SSS variability (monitored at threefixed mooring stations in the fjord) annual mean river dis-charges atmospheric oscillation patterns (state of the NorthAtlantic Oscillation ndash NAO) and nutrient availability whichall together account for changes in the fjord ecological state(Sakshaug and Myklestad 1973 Ottersen et al 2001) Weultimately apply the now standard modern analogue tech-nique (MAT) to the downcore distribution of dinocysts andtest its ability to reconstruct quantitatively a series of sea-surface parameters (SST SSS and primary productivity (PP))

in this fjord setting in order to establish a solid basis forfuture investigations of paleoclimate and paleoceanographicvariability

2 Physical settings and modern hydrology

The Trondheimsfjord extends from 6340prime N0945prime E at Oslashr-land at the fjord entrance to 6445prime N1130prime E at the fjordhead at Verdal (Boslashe et al 2003) and runs over into the Beit-stadfjord from Skarnsund 6350prime N1104prime E until Steinkjerat 6300prime N1128prime E (Fig 1) The fjord is 140 km long withan overall volume of 235 km3 and a surface area of 1420 km2The entire fjord area is divided into three basins the SeawardBasin the Middle Fjord and the Beitstadfjord The basins areseparated by three sills located at the fjord entrance (195 mwater depth) at the island of Tautra (98 m water depth) andat Skarnsund (140 m water depth) (Fig 1) (Jacobson 1983Ellingsen 2004) Sediment deposition in the Trondheims-fjord is largely controlled by the prevalent currents and theassociated pelagicmarine sediment load the topography andrivers delivering large amounts of terrigenous sediment intothe fjord (Boslashe et al 2003 Faust et al unpublished dataMilzer et al 2013) Bottom currents and tidal currents inthe subsurface layers prevent the rapid deposition of the fineparticle fraction in topographically elevated areas like sillsInstead the fine particle fraction primarily accumulates inthe more sheltered fjord basins Sedimentation rates are thushigher in the Middle Fjord and Beitstadfjord than at the fjordentrance and in the Seaward Basin where several currentspass at high frequencies (Fig 1) (Boslashe et al 2003 Milzeret al 2013) Silt to sand microfossils including dinocystand foraminifera accordingly mirror a local averaged patternwith however clear distribution gradients from the entranceto the inner part of the fjord (Milzer et al 2013)

21 Water mass exchange with the Norwegian Sea

The hydrology of the Trondheimsfjord is characterized bya three-layer system mainly driven by density differencesrelated to the seasonal temperature and salinity variabilityof the water masses inside and outside the fjord (eg Sak-shaug and Myklestad 1973 Stroslashmgren 1974 Jacobson1983) The bottom water is generally renewed twice a yearin FebruaryndashJune and SeptemberndashNovember by NAC andNCC waters respectively (Sakshaug and Myklestad 1973Stroslashmgren 1974 Jacobson 1983) The depth of the NACand its ability to cross the sill at the Trondheimsfjord en-trance depends on the depth of the overlying wedge-shapedNCC which in turn is linked to the prevailing seasonal windpatterns in the Nordic Seas (Jacobson 1983 Saeligtre 19992007) In early winter-spring when the NCC is supplied bylarge amounts of fresh water from the Norwegian hinterlandthe current is deflected offshore by northerly winds form-ing a shallow surface layer NAC waters replace the NCC

Clim Past 10 305ndash323 2014 wwwclim-pastnet103052014

G Milzer et al Dinocysts in the Trondheimsfjord during the past 25ndash50 yr 307

Fig 1 (A) Surface circulation in the North Atlantic region (EGC East Greenland Current EIC East Icelandic Current IC Irminger CurrentNAC North Atlantic Current NCC Norwegian Coastal Current)(B) bathymetric map of the Trondheimsfjord modified from Milzer etal (2013) Yellow dots indicate the core locations black dots the fixed mooring stations Roslashberg Ytteroslashy and Beitstad referred to in the text(C) temperature structure of the water column along the fjord axis in April 2011 (modified from Milzer et al 2013) locations of MC 99-1and MC S4-1 (black arrows) and the mooring stations Roslashberg Ytteroslashy and Beitstad (white triangles) are indicated along the fjord axis Thelocation of MC 61-1 is outside the cross section

at lower depths cross the sill at the fjord entrance and renewthe fjord bottom water In late summerearly autumn south-westerly winds induce a downwelling of the NCC inhibit-ing the crossing of NAC waters over the sill ConsequentlyNCC waters substitute the fjord water at intermediate depth(eg Sakshaug and Myklestad 1973 Jacobson 1983)

The intermediatesubsurface and surface waters in theTrondheimsfjord are subject to a two-layer estuarine circula-tion induced by large amounts of fresh water supplied by the6 major rivers Orkla Gaula Nidelva Stjoslashrdalselva Verdal-selva and Steinkjerelva (Fig 1) with a mean annual runoff of725 m3 sminus1 and up to 6430 m3 sminus1 during spring (Ellingsen2004 and references therein Pettersson 2012) Tidal surfacecurrents are considered to be the main force for internal mix-ing with a semidiurnal sea level change of ca 18 m (Jacob-son 1983) The tidal currents flow anti-clockwise enteringthe fjord on its southern edge during flood and flowing outalong the northern edge during ebb (Jacobson 1983)

22 Surface water characteristics over the past 50 yr

The temperature and salinity structures as well as the nutrientcontent in the surface waters vary seasonally associated withcoastal water exchange riverine input and internal mixing(Sakshaug and Myklestad 1973 Stroslashmgren 1974 Jacobson

1983) These processes in turn are to a large extent con-trolled by the prevalent winter NAO pattern of atmosphericvariability (Ottersen et al 2001) During a positive NAOphase the Norwegian climate is dominated by marine warmand wet air masses from lower latitudes causing strong pre-cipitation and continental runoff (eg Hurrell and van Loon1997 Ottersen et al 2001) During a negative oscillationstate Norway is affected by pulses of cold and dry conti-nental air masses resulting in extremely low temperaturesand reduced continental runoff This atmospheric influenceexplains to a high extent the modern mean annual SST andSSS variability over the Norwegian margin and the adjoin-ing coastal systems (Blindheim et al 2000 Visbeck et al2003) Contrary to the deep water layers which are largelydecoupled from the surface layers after the bottom waterexchange SSTs and SSSs strongly vary seasonally and an-nually in the Trondheimsfjord (Jacobson 1983) Accordingto instrumental measurements obtained from mooring sta-tions within the Trondheimsfjord (Fig 2) SSSs and SSTs at10 m water depth display a minor long-term salinity decrease(lt 05 PSU) and temperature increase (ca 1C) from 1963to 2005 (data from Trondhjem Biological Station of the Nor-wegian University of Science and Technology (NTNU)httpwwwntnuedubiologytbs) On an annual to decadal scale

wwwclim-pastnet103052014 Clim Past 10 305ndash323 2014


Fig 2 Hydrological and climatic data at mooring stations Roslashberg Ytteroslashy and Beitstad at 10 m water depth from 1963 to 2005 with(a)red lines 15 pt running average of monthly temperature data(b) green lines 7 pt running average of monthly salinity data(c) greyshaded graphs annual running average of river discharge rate anomalies (m3 yrminus1 in belowabove the 50 yr mean value)(d) grey dashedlines standard deviation of river discharge rates in (e) grey black-framed columns NAO index (Hurrell 1995) blue columns negativeanomalies of river discharge

mean annual river discharges (data from the Norwegian Wa-ter Resources and Energy Directorate (NVE)wwwnveno)are lowest when the NAO is in its negative state and highestduring a positive oscillation phase This NAO-forced vari-ability in river discharges is concomitant with the SST andSSS fluctuations with increased consistency towards the in-ner fjord (Fig 2) Calculated correlation coefficients betweenthe NAO index river discharge SST and SSS are how-ever low aroundR2

= 03ndash06 (p lt 005) with randomlyhigher values during spring and autumn aroundR2

= 06ndash09 (p lt 005) Annual to multi-annual SST changes corre-spond best with the NAO pattern in winter (with maximumvalues aroundR2

= 06p lt 005) suggesting a major influ-ence of atmospheric temperatures on the thermal signatureof the surface waters These low correlation factors can beexplained by the complex and often indirect ecological andenvironmental responses to NAO changes which may resultin relatively low values despite a proven impact (Ottersenet al 2001) Nevertheless the relationship between NAOphases river discharges SST and SSS changes seems to bebest established before 1980 This might be explained eitherby the long-term negative NAO phase until ca 1982 whichimposed a higher climatic stress on the fjord surface wateror alternatively by the implementation of local hydropowerplants in the early 80s such as in the Orkla river which mod-ifies the river input (LrsquoAbeacutee-Lund et al 2009) (Fig 2)

A proper investigation of the NAO impact (and associatedprecipitation and atmospheric temperature changes) on thequantity and type of nutrients in the fjord surface mixed layeris not possible yet due to the lack of time series of nutri-ent concentrations Short-term observations by Sakshaug and

Myklestad (1973) in 197071 however showed that nutrientssuch as nitrite nitrate orthophosphate and silicate fluctuateseasonally according to the depth and stability of the pycno-cline and riverine input The supply of NO3ndashN and PO4ndashP tothe Seaward Basin from continental runoff and rivers roughlyamounted to 12ndash70 t dayminus1 and below 015 t dayminus1 respec-tively with maxima in late spring and autumn Sewage sup-plied directly to the outer fjord yielded ca 25 and 075 t dailyof total N and total P respectively (Sakshaug and Myklestad1973)

3 Material and methods

Three sediment cores were retrieved along the fjord axis inStjoslashrnfjord at the fjord entrance in the central Seaward Basinand in the Middle Fjord (Fig 1 Table 1) with a multi coreras part of a research cruise of the research vesselFF Seismafrom the Geological Survey of Norway (NGU) in April 2011The sediment cores were sub-sampled on board at 1 cm inter-vals and split into two halves later in the lab one for radio-metric dating and one for dinocyst census counts The sed-iment samples were freeze-dried in order to save as muchmaterial as possible for dinocyst census counts

31 Chronology and sediment accumulation rates

Activities of 210Pb 226Ra 137Cs and232Th were measuredin ca 4ndash5 g of dried bulk sediment using a low back-ground high-efficiency well-shapedγ detector (Schmidt etal 2013) Activities are expressed in mBq gminus1 and errorscorrespond to 1 SD counting statistics The detector was



Table 1Site locations water depths and core lengths

Longitude Latitude Water CoreMulti-core Location [ East] [ North] depth [m] length (cm)

MC 61-1 Stjoslashrnfjord 9 86 6371 225 12MC 99-1 Seaward Basin 1019 6348 504 24MC S4-1 Middle Fjord 1098 6373 420 40

calibrated using IAEA standards (RGU-1 RGTh SOIL-6)Excess of210Pb (210Pbxs) was calculated as the difference ofthe total activity of210Pb in the sediment and the supportedactivity by its parent isotope226Ra The error on210Pbxstherefore corresponds to the propagation of errors in mea-sured210Pb and226Ra activities

The dating of the multi-cores relies on210Pb a natu-rally occurring radionuclide in the sediment with a half-life of 223 yr This method is commonly used to estimateshort-term (years to decades) sediment accumulation ratesin continental and oceanic environments for the last 40 yr(Appleby 2001) Briefly dating is calculated using the ex-cess activity of210Pb (210Pbxs) which derives from atmo-spheric fallout and water column scavenging and is rapidlyincorporated into the sediment The unsupported210Pb in-corporated into the sediment decays with time according toits half-life In order to estimate ages andor sediment ac-cumulation rates several models have been developed as-suming different hypotheses such as constant initial concen-tration (CIC model) constant rate of supply (CSR model)or constant flux and constant sedimentation (CFCS model)(Kirchner 2011 among others) In this study we applied theCFCS model and examined its accuracy in dating of thesedimentary archives using137Cs (T12 = 30 yr) (eg Rob-bins and Edgington 1975 Durantou et al 2012) This artifi-cial radionuclide serves as an independent time-stratigraphicmarker due to elevated activity peaks associated with nuclearweapon test fallouts (maximum in 1963) and the Chernobylaccident in 1986 (Durantou et al 2012)

32 Dinocysts sample preparation and ecologicalbackground

Dinoflagellate cysts were retrieved from thelt 150 microm frac-tion The treatment of the samples follows a standard sam-ple preparation procedure (eg Stockmarr 1971 de Vernal etal 1996) slightly modified at EPOCUniversiteacute Bordeaux 1(Penaud et al 2008httpwwwepocu-bordeauxfrindexphplang=framppage=eq_paleo_pollens) The samples weretreated with cold hydrochloric acid (HCl) (10 25 and 50 )and cold 40 and 70 hydrogen fluoride (HF) to dissolvecarbonates and silicates and washed through a 10 microm nylonmesh The residue was centrifuged and additionally swirledin a large watch glass to remove high amounts of amor-phous organic material (AOM) (Riding and Kyffin-Hughes

2004 Milzer et al 2013) The final residue was mountedwith glycerine jelly between microscope slides On average300 cysts per slide were counted using a Zeiss Axioscopelight microscope at 40times magnification Absolute abundances(cysts cmminus3) of dinocysts and palynomorphs were calcu-lated using theLycopodiummarker grain method (Stockmarr1971) Relative abundances () of each species were calcu-lated including only dinocysts

The taxonomical nomenclature corresponds to Rochon etal (1999) and Head et al (2001 2006)Operculodinium cen-trocarpum sl includes the long- and short-process formsof O centrocarpum sensu(Wall and Dale 1967) andOcentrocarpumndash Arctic morphotype (de Vernal et al 2001)Dinocyst species within theSpiniferites SelenopemphixIslandinium Brigantediniumand Impagidinium taxa weregrouped according to Marret and Zonneveld (2003) as sum-marized in Milzer et al (2013) The resulting species rich-ness in the studied sediment cores from the Trondheims-fjord amounted to a maximum of 17 dinocyst taxa and 4Chlorophycean and Prasinophyceae species The spatial dis-tribution of the dominant species in surface sediments ofthe Trondheimsfjord was recently discussed by Milzer etal (2013) The following section summarizes the ecologyand environmental background of each recognized speciesUnless otherwise specified most information is taken fromMatthiessen (1995) Rochon et al (1999) Marret and Zonn-eveld (2003) de Vernal and Marret (2007) van Nieuwenhoveet al (2008) and Zonneveld et al (2013)

Pentapharsodinium daleiis associated with cold and strat-ified waters as typical for Norwegian fjords and other tem-perate and polar embayments The species belongs to theautotrophic taxa and generally blooms in springearly sum-mer when surface waters contain high amounts of nutrientsfrom previous winter mixing The autotrophic speciesOp-erculodinium centrocarpumis considered as a cosmopoli-tan species which is able to adapt quickly to environmen-tal changes and is therefore often observed in coastal watersandor at the coastaloceanic boundary (Wall and Dale 1967Dale et al 2002 1999b) In the Nordic Seas this speciesis associated with the pathway of the warm and saline wa-ters carried by NAC (de Vernal et al 2000)Selenopem-phix quantasl is a heterotrophic species which has beenmainly documented from subtropical coastal areas in re-gions characterized by mesotrophic to eutrophic conditionsThe abundance of the heterotrophic speciesBrigantedinium



spp largely depends on the presence of diatom preys whoseoccurrence is similarly linked to nutrient enrichment inthe photic layer in various environments (eg regions ofenhanced upwelling sea-ice edge zones etc) The het-erotrophic taxonIslandinium minutumshows a bipolar dis-tribution in coastal areas north of 30 N and south of 30 Swithstanding large (inter-)seasonal fluctuations of tempera-ture salinity insolation nutrient concentration and long sea-ice cover extension Cysts ofLingulodinium machaeropho-rum are reported to be restricted to regions with summertemperatures above 10ndash12C An enhanced occurrence ofLmachaerophorumin southern Norwegian fjords has been re-lated to eutrophication due to industrial and agricultural ac-tivity (Saeligtre et al 1997 Dale et al 1999a) and along othercoastal areas to increased freshwater advections (Zaragosiet al 2001 Lezine et al 2005)Ataxiodinium choanumisa coastal to open oceanic fully marine species with high-est relative abundances in eutrophic environments (Marretet al 2004) In this study the species is most likely re-lated to an enhanced influence of Atlantic-derived watersCysts ofNematosphaeropsis labyrinthusare generally linkedto outer neritic to oceanic environments and to eutrophicconditions within the North Atlantic ocean (Devillers andde Vernal 2000) TheSpiniferitestaxa are dominated bySpiniferites ramosusandSpiniferites elongatus Spiniferitesramosussl is considered as a cosmopolitan species linkedto upwelling or a well-mixed surface water layer resultingin mesotrophic to eutrophic conditions Nowadays highestoccurrence ofS elongatussl is found around Iceland po-tentially indicating a preference to colder and lower salinitybut eutrophic surface water conditions (Marret et al 2004)Cysts ofSpiniferites lazus Spiniferites mirabilis SpiniferitesmembranaceusandSpiniferites delicatusare associated withmarine warm and saline waters in oligo-mesotrophic envi-ronmentsS bentoriiis associated with coastal high produc-tivity environments and has been found in Saanich Inlet (BCCanada) Since the total abundance of theseSpiniferitestaxain the Trondheimsfjord does not exceed 3 we relate theirpresence to a stronger marinecoastal influence in surfacelayers assigned as the combined termSpiniferites ldquoAWrdquowithout distinguishing their particular species-specific char-acteristics All species grouped withinImpagidiniumsppgenerally thrive in fully marine environments (rarely neritic)with oligotrophic to eutrophic conditions The distribution ofBitectatodinium tepikienseis related to the transitional zonebetween the subpolar and temperate zones in the Norwe-gian Sea (Groslashsfjeld and Harland 2001) Higher abundancesof this species may indicate a strong seasonal temperaturegradient with enhanced surface water stratification Cysts ofPolykrikos schwartziiare generally found in sediments fromsubtropical to temperate continental shelves estuaries andshelf seas throughout the world except for polar seas char-acterized by a high nutrient level as well as turbulent wa-ters (Matsuoka et al 2009) Cysts ofAlexandrium tamarenseare reported in coastal sediments of temperatesubtropical re-

gions with brackish to fully marine oligotrophic to eutrophicenvironments from the coasts of eg Germany Sweden andSpain (Nehring et al 1997 and references therein) The in-creased abundance ofHalodinium minor Radiosperma corb-iferum and Hexasteria problematicais related to moderatesalinity and high nutrient content in surface waters in tran-sitional brackish to marine environments (eg Parke andDixon 1964 McMinn 1991 Matthiessen 1995 Price andPospelova 2011 Sorrel et al 2006)

33 Multivariate analyses

Multivariate analyses were applied to the cyst assemblagesin order to extract major patterns in species abundances andto investigate their spationdashtemporal organization accordingto the various environmental settings of the studied sedimentcores The analyses were conducted using the R Project freesoftware package for statistical computing version 300 (RDevelopment Core Team 2008httpwwwr-projectorg)and paleontological statistics software ldquoPASTrdquo (httpfolkuionoohammerpast) by Hammer et al (2001) Since drybulk density measurements were only conducted in one core(MC 99-1) it was not possible to calculate the species accu-mulation rates in all cores based on the cyst concentrationsAll statistical analyses were therefore conducted using therelative cyst abundances only ie they refer to the composi-tion of the assemblages but not to the rate of species accumu-lation (flux) at the sediment water interface The statisticallybased classification of the investigated sediment samples wascarried out on both the separate data sets related to singlesediment cores and the combined cyst assemblages obtainedfrom the three studied multi-cores The latter were analysedusing correspondence analysis (CA) in order to constrainsamples (years) in terms of characteristic species and theirassociated ecological environmental and stratigraphic con-texts in the Trondheimsfjord (Hammer et al 2001) How-ever differences between the samples and the core sites arerelatively small We therefore additionally performed a clus-ter analysis on the data set of every single sediment core us-ing the BrayndashCurtis dissimilarity index to more objectivelyhighlight the differences of the assemblages

34 Quantitative reconstructions of the sea-surfaceparameters SST SSS and PP

The modern analogue technique (MAT) was applied to therelative dinocyst abundances in the studied sediment coresfor quantitative reconstructions of winter and summer SSSand SST and annual primary productivity (PP) (de Vernalet al 2001 Radi and de Vernal 2004 Guiot and de Ver-nal 2007) The MAT method was successfully developedfor the reconstruction of the sea-surface parameters in openocean conditions for most of the existing set of marine mi-crofossils (eg Imbrie and Kipp 1971 Birks 1995 Maslinet al 1995 Pflaumann et al 1996 Kucera et al 2005



Waelbroeck et al 2005 Guiot and de Vernal 2007 Telfordand Birks 2011) The application of MAT in this work isconsidered as an initial test of the potential and the qual-ity of dinocyst-based reconstructions of sea-surface parame-ters in fjord environments using transfer functions We there-fore strictly followed the standard procedure according tode Vernal et al (2001) and Guiot and de Vernal (2007) us-ing one of their most recently validated n= 1207 databases(see the successive steps in the construction of the mod-ern database athttpwwwgeotopcafrbases-de-donneesdinokysteshtml see also Radi and de Vernal 2008 Radiet al 2009 Bonnet et al 2009) as summarized as followsHydrographic (summer and winter SST and SSS) and an-nual primary productivity data at 0 m water depth whichwere averaged over the period 1900 to 2001 (NODC 2001)and 2002 to 2005 (VGP model applied to MODIS chloro-phyll data) respectively were assigned to each of the ref-erence modern sites (ie analogues) The sea-surface condi-tions were reconstructed from the fossil records with regardto the similarity to the modern dinocyst spectra In order todetermine the best number of analogues to retain with regardto the estimated values and the associated degree of accuracywe conducted MAT with different numbers of analogues (Ta-ble 2) The maximum weight is given for the closest ana-logue in terms of statistical distance and the associated rangeof the environmental parameters Best results for the sea-surface parameter reconstructions and the associated errorswere achieved using 5 analogues as previously tested withthe other databases (see de Vernal et al 2000 2001) Thischoice was justified after several tests done to determine thebest number of analogues according to a compromise regard-ing the degree of accuracy of estimations (Table 2) A mini-mum number of analogues (3) minimized the error bar on re-constructions but is not statistically robust whereas a numberof analoguesgt 5 (eg 10) provided an artificially high errorbar Reconstructed values based on the assemblages from thedifferent core locations using 3 5 and 10 analogues also didnot reveal large differences but in turn confirmed the use of5 analogues for the estimation of the sea-surface parametersas a relative meancompromise The associated estimated pa-rameters are presented as the weighted averaged values andthe related minimum and maximum values eg SST minSST and max SST found in the 5-analogue set (Table 3)The accuracy of the approach is assessed by the correlationcoefficient between observed and estimated values for eachreconstructed parameter when the test is performed on thecalibration data set (ie the modern database excluding thespectrum to analyse see de Vernal et al 2001 2005) Thisstep furthermore provides residual differences between ob-servations and estimates from which is derived the root meansquare error (RMSE) equivalent to the standard deviation ofthese residuals (eg de Vernal et al 2001 2005 Radi and deVernal 2004 Guiot and de Vernal 2007) Additionally wetested our data for the risk of autocorrelation by running theMAT with the reference data sets without (1) the identified

analogues sites in the Norwegian Sea and (2) sites with thehighest recall ratio

The MAT was run using R Project for statistical com-puting version 300 (R Development Core Team 2008httpwwwr-projectorg) and a script developed by JGuiot in the BIOINDIC package (httpswwweccorevfrspipphparticle389) We calculated the reconstructions ofwinter and summer SSTs and SSSs as well as PPs us-ing the original data set of Radi and de Vernal (2008)collected in the northern hemisphere of the North At-lantic Arctic and North Pacific oceans with additionaldata points (database available from the DINO9 work-shop httpwwwgeotopcauploadfilesbabillardcongres_et_ateliersHandout20Dino9-Workshoppdf) bringing thetotal number of sites to 1207 With this database mostaccurate reconstructions (based on linear regressions ofestimated vs observed data) can be achieved for winterand summer SSTs withR2 = 095 and RMSEP= 11CandR2 = 096 and RMSEP= 149C respectively SummerSSSs similarly reveal a good accuracy withR2 = 074 andRMSEP= 240 PSU The lowest accuracy was achieved forwinter SSSs withR2 = 069 and RMSEP= 230 PSU Thevalidation test for the reconstruction of primary productiv-ity yields a good accuracy withR2 = 084 and quite a largeRMSEP= 5458 g C mminus2 yrminus1 (Table 2)

In a final step we compared separately the reconstructedvalues with the associated measured time series monitored at10 m water depth at the respective mooring stations (Roslashbergfor core MC 61-1 and MC 99-1 Ytteroslashy for core MCS4-1) and tested the similarity using the MannndashWhitney(Wilcoxon) test (Bigler and Wunder 2003 von Detten et al2008)

4 Results and discussion

41 Chronologies and sediment accumulation rates

210Pbxs activities in MC 99-1 in the Seaward Basin rangefrom 37 to 159 mBq gminus1 decreasing downcore according tothe radioactive decay of210Pb with time (Fig 3) With re-gard to the CFCS model the sediment accumulation ratein MC 99-1 is 049 cm yrminus1 and encompasses the last 49 yrThe counting error increases downcore reaching up to max35 yr (Fig 3) The137Cs profile in MC 99-1 shows remark-able peak activities with maximum values of 53ndash59 mBq gminus1

at 12ndash13 cm and of 36ndash37 mBq gminus1 at 23ndash24 cm sedimentdepths According to the210Pbxs-based dating sedimentsat these depths correspond to an age of 1986plusmn 17 yr andca 1963ndash1965plusmn 34 yr respectively which is in good agree-ment with the known nuclear power plant accident in Cher-nobyl (1986) and the nuclear weapon test fallouts (max in1963) (Fig 3)

Activities of 210Pbxs at the fjord entrance (core MC 61-1)decrease continuously with depth from 132 to 69 mBq gminus1



Table 2Results of validation test for the reconstructions of the winter and summer sea-surface temperature (SST) parameters salinity andprimary productivity based on the database comprising 1207 reference sites (see Sect 34) with 3 5 and 10 analoguesR2 is the correlationcoefficient between observed and estimated values The RMSEP (Root Mean Square Error of Prediction) is calculated by dividing thedatabase into calibration and verification data sets1 mean refers to the mean of the ranges between the estimated minimum and maximumvalues

MAT_5 analogues MAT_3 analogues MAT_10 analogues

Environmental parameters R2 RMSEP 1 mean R2 RMSEP 1 mean R2 RMSEP 1 mean

Winter SST 095 105 197 098 106 125 097 109 313Winter SSS 069 23 270 067 235 182 071 221 399Summer SST 096 149 310 096 150 202 096 153 478Summer SSS 074 24 339 073 246 224 075 238 508Primary productivity 084 5458 10009 083 5609 6467 083 3576 15185

The deepest layer of core MC6 1-1 (105 cm) shows an in-crease in activity (82 mBq gminus1) with a concomitant increasein 232Th (data not shown) The long-lived232Th occurs natu-rally and is usually associated with the detrital fraction Vari-ations in232Th activities are thus assumed to trace changes inthe detrital fraction proportion (Bayon et al 2009 and refer-ences therein) Sedimentation rate in MC 61-1 was thereforeestimated based on210Pbxs activities normalized by the mea-sured232Th activities to correct for any deviations related tosubstrate changes (Fig 3) The corresponding sediment accu-mulation rate yields 046 cm yrminus1 The 12 cm-long sedimentcore thus comprises the past 25 yr (plusmn max 11 yr) The timeperiod encompassed in this sedimentary archive is too shortto reveal peak activities associated with the Chernobyl ac-cident or past nuclear weapon tests The absence of activitypeaks in the137Cs profile however serves as an indirect val-idation (Fig 3)

The activities of210Pbxs in MC S4-1 in the Middle Fjordrange from 44 to 11 mBq gminus1 from the sediment surfaceto the lowermost depths with fairly low activities from 11to 16 mBq gminus1 at depths deeper than 28 cm (Fig 3) Thecore is located in a deep basin surrounded by hills and theriver Verdalselva discharges significant amounts of fresh wa-ter and terrigenous sediments from the hinterland into theMiddle Fjord We assume that relatively low210Pbxs ac-tivities as well as the changes in the activity profile below28 cm sediment depths may be related to changes in sedimentsource and delivery We therefore normalized the measured210Pbxs activities with the depth-related activities of the de-trital tracers232Th similarly to the chronology in MC 61-1The sedimentation rate in the 38 cm-long multi-core MC S4-1 yields 068 cm yrminus1 and consequently encompasses the last55 years with a maximum error ofplusmn5 yr (increasing down-core) Peak activities of137Cs (54 mBq gminus1) at depths of 16ndash18 cm (according to the chronology accomplished by cor-recting for changes in the detrital fraction) correspond to theyears 1985ndash1987 and can thus be attributed to the Chernobylaccident in 1986 (Fig 3) At depth 31ndash32 cm the expectedpeak related to fallout tests is not well expressed However

the occurrence of detectable137Cs activities testifies to sedi-ments deposited after 1950

42 Dinocyst concentrations and preservation

Palynomorphs identified within the studied cores aremainly dominated by dinocysts whereas Chlorophyceanand Prasinophyceae algae rarely account for 5 of the to-tal palynological assemblages (excluding pollens) The fol-lowing information will therefore only relate to dinocystsMaximum absolute abundances of dinocysts were foundin MC 61-1 in Stjoslashrnfjord at the fjord entrance fluctuat-ing between ca 10 000 and 130 000 specimens cmminus3 withmean concentrations of 72 000 specimen cmminus3 (Fig 5) Val-ues were significantly lower in the cores located in theMiddle Fjord and the Seaward Basin with mean dinocystconcentrations of 25 000 specimens cmminus3 (MC S4-1) and14 000 specimens cmminus3 (MC 99-1) (Figs 6 7)

These relatively high mean concentrations correspondwell to observations by Dale (1976) from a limited set ofsurface samples and our own analysis based on 59 surfacesamples evenly distributed in the Trondheimsfjord (Milzeret al 2013) as well as to concentrations observed in innershelf and fjord environments along the Norwegian coast byGroslashsfjeld and Harland (2001) Differences in mean cyst con-centrations among the studied multi-cores are most likelymainly related to dilution linked to the current circulationfeatures and to the proximity of river inlets and associatednutrient and terrigenous sediment loads Highest absolutecyst abundance and largest species evenness are observedin MC 61-1 in Stjoslashrnfjord at the fjord entrance and can berelated to the proximity of the Norwegian Sea Stjoslashrnfjordis easily flushed with Atlantic water and is likely to ex-perience a periodical coastal upwelling resulting in an en-hanced nutrient availability Additionally the dilution of ma-rine biogenic material by terrestrial organic and lithogenicparticles might be limited (sediment accumulation is lowestat site MC 61-1 (046 cm yrminus1) due to the absence of ma-jor river inlets in Stjoslashrnfjord contrary to the inner parts of



Fig 3 Core chronology of MC 61-1 (centre) at the fjord entranceMC 99-1 (top) in the Seaward Basin and MC S4-1 (bottom) in theMiddle Fjord showing(A) 210Pbxs or 210PbxsTh (ie 210Pbxs nor-malized to232Th explanation in the text) profiles(B) 137Cs pro-files (except for MC 61-1) and(C) calculated ages based on theCFCS model The arrows indicate the expected position of the137Cs peaks related to maximum weapon test atmospheric falloutand the Chernobyl accident according to the chronology based on210Pbxs

the Trondheimsfjord The Seaward Basin receives dischargesfrom 4 of the 6 major rivers which end in the Tronsheims-fjord (Fig 1) Associated nutrient and terrigenous loads aswell as dinocysts thriving in the surface layers are distributedacross the fjord linked to the general circulation pattern ofthe surface currents during flood tide (Jacobson 1983) Sed-iment (sedimentation rate in core MC 99-1 049 cm yr1) andcyst accumulation in the deep Seaward Basin is thus lim-ited despite the proximity of major river inlets Moreoverthe observed relatively low cyst concentrations measured atsite MC 99-1 (mean= 14 000 cysts cmminus3) can also be ex-plained by dilution by fine sediments carried into the fjordbasin during bottom water renewal events

Fig 4 Correspondence analysis based on relative cyst abundancesof all cores combined as one data set The species are ordinated bythe CA according to abundance and downcore variability of cystsThe coloured numbers are organized by sediment cores MC 61-1 blue MC 99-1 red and MC S4-1 green samples belonging togroup I are placed on the positive axis 1 in correspondence with thedominance ofP dalei minimum abundances or absence of othertaxa samples categorized into group II are placed mainly in thecentre of the CA and are characterized by a successive decrease inP daleiand consequently higher abundances of most of the otherspecies Samples representative of group III with the highest cystevenness are located on the far negative axis 1

The Middle Fjord is more sheltered than the SeawardBasin with relatively high riverine (and nutrient) input andhigh sedimentation rates (MC S4-1 068 cm yrminus1) Cyst con-centrations are nevertheless relatively low and we assumethat cyst assemblages in the Middle Fjord are subject to dilu-tion due to the large sediment input from the hinterland

Ultimately the preservation ability against post-depositional degradation due to sediment oxygenationmay alter the original cyst assemblages (Zonneveld et al1997 2001) The dominant species in the TrondheimsfjordP dalei is considered to be relatively resistant to aerobicdecay in contrast to the moderately sensitive cysts ofOcentrocarpum L machaerophorumand the highly sensitiveBrigantediniumspp species (Zonneveld et al 1997 2008)Post-depositional degradation may thus contribute to therelatively poor cyst evenness in the fjord sediments Still therelatively high sedimentation rates might limit degradationprocesses by minimizing the time organic remains areexposed to oxygen in the bottom and pore waters (Blume etal 2009)



Fig 5 Cyst assemblages in MC 61-1 and associated groups according to cluster analysis cyst concentrations and Chlorophycean andPrasinophyceae algae are given in nb ccminus1 relative dinocyst abundances are given in the NAO index is shown to the right (Hurrell 1995)

Fig 6 Cyst assemblages in MC 99-1 and associated groups according to cluster analysis cyst concentrations and Chlorophycean andPrasinophyceae algae are given in nb ccminus1 relative dinocyst abundances are given in the NAO index is shown to the right (Hurrell 1995)Black bar on the right axis marks the period (1978ndash1985) of major hydropower developmentimplementation in the Orkla river (LrsquoAbeacutee-Lundet al 2009)

43 Species diversity and temporal succession ofassemblages

Species richness and distribution in the three investigatedsediment cores follow the general abundance patterns de-scribed by Milzer et al (2013) from surface sediments in theTrondheimsfjord Dinocyst assemblages are dominated byPdalei andO centrocarpumMaximum mean relative abun-dances ofP dalei were found in MC 99-1 in the SeawardBasin (67 ) (Fig 6) and lowest values in MC 61-1 (50 ) atthe fjord entrance (Fig 5) Cysts ofO centrocarpumshowedthe highest mean relative abundances in MC 61-1 (34 )

followed by MC 99-1 and MC S4-1 with 25 (Figs 6 and7) Cysts ofS quantaand Brigantediniumspp are subor-dinate in all multi-cores both averaging 5 in MC 61-1ca 17 in MC 99-1 and 36 (S quanta) and 51 (Brig-antediniumspp) in MC S4-1 All other species do not exceedgt 3 of the total assemblage

The recorded species richness and abundances are compa-rable with the assemblages identified in modern settings ofcoastal Norway (Groslashsfjeld and Harland 2001 Groslashsfjeld etal 2009) with the exception ofP daleiandO centrocrapumThe reverse dominance of these species in the Trondheims-fjord with P dalei being the most abundant species may



Fig 7 Cyst assemblages in MC S4-1 and associated groups according to cluster analysis cyst concentrations and Chlorophycean andPrasinophyceae algae are given in nb cmminus3 relative dinocyst abundances are given in the NAO index is shown to the right (Hurrell1995)

be explained by the hydrological differences between thewell-stratified brackish and sheltered fjord and the dynamiccoastal environment as already suggested by Groslashsfjeld andHarland (2001) In order to determine precisely changes inthe assemblages to the ambient environmental parametershowever there is a need for further information such as tem-poral variations in grain size total organic carbon or amountsof nutrients which are not available yet

The cyst assemblages of the three multi-cores show a highdegree of temporal and spatial similarity This is particularlyevident in the results of the cluster analysis with similar-ity indices gt 07 The classification of the downcore vari-abilities by CA mainly relies on the varying relative abun-dances of the dominant and the subordinate speciesP daleiO centrocarpum S quantaandBrigantediniumspp Statis-tical analysis with reduced weight on the dominant taxa us-ing log-ratio exclusion of the dominant species or based onraw counts or concentration values did not allow for a sta-tistically based distinction of the samples and thus associa-tion with ecological or environmental parameters Based onCA of the relative cyst abundances we however identifiedthree main groups constituting the temporal and spatial orga-nization of the combined data set of cyst assemblages Clus-ter analysis performed on the cyst assemblages of the singlemulti-core assemblages confirms the temporal distribution ofthe three groups defined by the CA

Group I is mainly represented in the 1960s and 1970s andtherefore only grasped in the MC 99-1 sedimentary archivesin the Seaward Basin and MC S4-1 at the fjord head (Fig 4)This group is characterized by a strong dominance ofP dalei(60ndash80 ) overO centrocarpum(lowest recorded abun-dance within all three cores) (Figs 6 and 7) All other taxa

mainly represented bySpiniferitessppS ramosus S elon-gatusandSpiniferitesldquoAWrdquo as well as A choanumandSquanta generally add up tolt 3 with very low variability

Group II mostly represented from around 1980ndash1982 ischaracterized by slightly decreased abundances ofP dalei(48ndash75 ) and in turn increased occurrences ofO cen-trocarpum(19ndash35 )S quanta(03ndash7 )Brigantediniumspp (03ndash9 ) as well as generally minor higher occurrenceand variability of all other taxa (Fig 4)

Group III is temporally distributed during ldquotransitional pe-riodsrdquo between groups I and II This group is characterizedby lowest relative abundances ofP dalei and highest oc-currences of most other taxa (Fig 4) and explains most ofthe dinocyst spectra in all three cores during relatively shorttime intervals of up to 4 yr during the 70s to 80s transition aswell as throughout the final part of the records from ca 2004(2000 in MC 61-1) (Figs 5ndash7)

According to the core chronologies the resolution of themulti-cores is ca 2 yr cmminus1 The chronological framework ofthe studied cores is related to the classical statistical count-ing error increasing downcore with a maximum deviationof 5 yr (in MC S4-1) Inconsistencies in the order of 2ndash3 yrin the timing of group changes between the three sedimentcores may thus be partly explained by the sampling strat-egy (1 cm-thick homogenized samples) and fall within theerror range of dating (higher in the bottom parts of the sed-iment cores) Still there are a few single samples or seriesof samples which fail to follow this general scheme pointingeither to some limitation in the statistical ordination of thedata sets or to local hydrologicalndashsedimentological settingscausing irregularities in cyst abundances and diversity Cystassemblages in MC 61-1 are largely affected by the adjoining



Norwegian Sea causing relatively enhanced species richnessand creating environmental conditions which are relativelydifferent from those affecting the other core locations Thedinocyst distribution in MC 99-1 is generally organized ac-cording to the main grouping Annual andor biannual peri-ods of relatively enhanced species variability and diversityrefer to group III and occur from 1977 to 1980 and around1996 (Fig 6) The classification of the cyst assemblage incore MC S4-1 (Fig 7) displays a series of annual to bi-annualshifts from group I to group III throughout the 70s to 80stransition as a result of minor changes in dinocyst abundancein the years 1976 1981 and 1987

The three groups and their temporal succession are asso-ciated with distinct environmental conditions and potentiallywith the history of development policies in the Trondheimarea in the 70s and the 80s and their related impact on thephysico-chemical structure and trophic state of the surfacewaters

Investigations in southern Norwegian fjords showed thathigh amounts of nutrients related to industrial activity and ef-fluents caused an increase in the cyst concentration and abun-dance of heterotrophic species such asL machaerophorumand in turn a sharp decrease in the contribution ofP dalei(Saeligtre et al 1997 Dale et al 1999a) Regulations to re-duce the impact of agricultural and industrial activities on thequality of Trondheimsfjord waters were successfully appliedaround 1980 (Tangen and Arff 2003) If eutrophication wereone of the important driving forces for the dinocyst produc-tion in the Trondheimsfjord during the past 50 yr this wouldresult in a temporal pattern of the assemblage distribution op-posite to the one observed in the present study (Figs 6 and 7)We therefore conclude with the exception of a potential mi-nor impact due to the implementation of hydropower plantsthat dinocyst abundance patterns over the last 50 yr were es-sentially controlled by natural environmentalclimatologicalchanges

The period from the 1960s and the 70s which is associ-ated with group I is characterized by a long-term negativeNAO phase with lower than normal river discharges intothe Seaward Basin and the Middle Fjord as well as rela-tively high mean SSSs and low mean SSTs (Fig 2) Thispattern was interrupted during the early 70s by a short in-terval of positive NAO pattern and respective changes in thephysico-chemical characteristics of the surface waters Thehigh representation of the cyst assemblages typical of groupI within cores MC 99-1 and MC S4-1 during the 60s and70s is supposed to result from increased continental influ-ence in the inner fjord basins which may intensify changes inthe sea-surface parameters associated with the NAO phasesLow amounts of river discharges from the hinterland andthus reduced supply of nutrients into the fjord impede theproduction of phytoplankton prey for heterotrophic speciesfrom spring to autumn and might explain the scarcity of taxasuch asS quanta BrigantediniumsppI minutumandLmachaerophorumin group I (Figs 6 and 7)

After 1980 the period is mainly represented by group IIDuring this period the NAO changed to a long-term posi-tive oscillation pattern resulting in generally higher SSTshigher than normal mean river discharges and lower SSSs(Fig 2) The temporal distribution of the cyst assemblagesreferred to group II within the three studied cores is how-ever less straightforward and homogenous than group I be-fore the 1980s (Figs 4ndash7) This may relate to ill-definedNAO conditions after 1996 leading to rapid fluctuations inSSTs river discharges and SSSs Anyhow regarding thegeneral homogeneity of the assemblages even minor devi-ations in the species abundance may lead to the determina-tion of the samples into another group and cannot neces-sarily be explained by measurable changing environmentalconditions Still we assume that the generally higher con-tributions of heterotrophic species and cysts with a prefer-ence for mesotrophic conditions within all cores after 1980are indicative of enhanced phytoplankton production relatedto increased nutrient load from rivers and internal mixingGroup III associated with the largest species richness gen-erally occurred throughout the 70s to 80s transition as wellas after 2000 both periods during which the NAO shiftedfrom one long-term mode to another Its absence from themost recent part of core MC 99-1 (Fig 6) might be relatedto the regulation of river inputs into the Seaward Basin dueto the implementation of hydropower plants in the early tomid 1980s (Orkla river LrsquoAbeacutee-Lund et al 2009) The artifi-cially modified river discharges might account for the hetero-geneous patterns of discharge anomalies between the riversOrkla Gaula and Nidelva (low discharges) and the othermajor tributaries to the Trondheimsfjord (high discharges)(Fig 2) Alternatively in the central Seaward Basin changesin the surface water characteristics due to enhanced riverinedischarge may be diminished by mixing with the incomingtidal currents and the ambient fjord surface water

Calculated correlation coefficients between the NAO in-dex and the dinocyst assemblages however yield low val-ues (R2 lt 05) These low values are likely to be explainedby the complex and rather indirect link between the NAOand the cyst assemblages translated by river discharges andthe water column stability Additionally as previously men-tioned downcore changes in cyst assemblages are minor andmay therefore be statistically not significant enough whencompared with long- and short-term high-amplitude changesin the NAO index

44 Quantitative reconstruction of SSTs SSSs and PP

The MAT applied for the reconstruction of SSTs and SSSsrelates to the 5 best analogues for every multi-core samplewith distances below the statistical thresholds for all coresTests for autocorrelation due to the analoguesrsquo sites did notreveal significant changes in the reconstructions yieldingmax 1R2

= 02 and a correspondingly increased RMSEPReconstructed temperatures and salinities accordingly yield



Fig 8 Winter and summer SST and SSS reconstructions based on cyst assemblages at every core site red line SST green SSS greysurrounding lines minimum and maximum estimates for each reconstructed value calculated from the set of the 5 selected analogues greydots instrumental temperatures and salinities measured at the respective mooring station Roslashberg for MC 61-1 and MC 99-1 and Ytteroslashy forMC S4-1 black lines are the three-point running average of the instrumental temperatures and salinities

a maximum difference ofplusmn2C3 PSU compared with thecomplete data set Hence although we cannot exclude therisk of spatial andor temporal autocorrelation in the recon-structed data our estimated values are relatively robust

The reconstructed hydrological values in all cores main-tain the overall seasonal temperature and salinity structureof the surface water as recorded at the respective mooringstations (Roslashberg for MC 61-1 and MC 99-1 and Ytteroslashyfor MC S4-1) with low SSTs and higher SSSs in winterand high SSTs and slightly lower SSSs in summer Gener-ally all estimated minimum values of summer and winterSSTs and SSSs as well as PP are significantly underestimatedin all cores compared with the instrumental measurementsat the respective mooring station (Wilcoxon testp lt 005)In MC 61-1 winter and summer SST and max SST rangebetween 32C and 66C and 124C and 174C re-spectively Summer salinity varies from 281 PSU (SSS) to328 PSU (max SSS) and winter SSS and max SSS from304 PSU to 335 PSU Winter SST and max SST in MC 99-1 yield 11 and 37C (Table 3 Fig 8) and a summer SSTand max SST 90C and 137C respectively Summer SSSand max SSS vary between 281 and 324 PSU and between306 and 335 PSU in winter In MC S4-1 in the Middle FjordSST and max SST range from 18C to 51C in winterand from 110C to 162C in summer SSS and max SSSyield 283 PSU and 324 PSU in summer and 306 PSU and333 PSU in winter respectively

Regardless of the site-specific chronological extent andtime resolution of the downcore records reconstructed sea-surface abiotic parameters in terms of maximum and highestprobability values are in relatively good agreement with the

modern instrumental data and general temporal variability(Table 3 Fig 8) However the correspondence of the esti-mated values to the instrumental measurements varies withthe core location the reconstructed sea-surface parameterand the reference sites identified At the fjord entrance (MC61-1) the best agreement with the instrumental measure-ments in the central Seaward Basin (Roslashberg) is achieved bythe reconstructions of the the summer SSTs and winter maxSSTs (Table 3 Fig 8) In MC 99-1 in the Seaward Basin esti-mated winter max SSTs and summer SSTs and max SSS areconsidered as significantly similar (p gt 005) to the instru-mental measurements Dinocyst-based reconstructions fromthe Middle Fjord reveal significant correspondence with theinstrumentally measured temperatures in Ytteroslashy only for thesummer SSTs with mean values of 110C and 104C andwinter max SSTs with 42C and 51C respectively

Still despite the relative good fit between the estimatedvalues and the instrumental measurements there are sev-eral aspects which may cause the observed deviations andortechnical biases which have to be carefully considered wheninterpreting the data The hydrographic data which constituteour reference modern data average the environmental infor-mation over several decades The dinocyst spectra may in-clude populations representative of several years to a mil-lennium depending on the sedimentation rate and the bi-ological mixing depth at the sampling sites (de Vernal etal 2001) The reference data sets may therefore not bestrictly representative of the modern environmental settingin the Trondheimsfjord regarding the resolution of the multi-cores Furthermore the parameters reconstructed ie win-ter and summer SSTs are not necessarily representative of



Table 3 Descriptive statistics (mean range minimum and maximum values) of hydrographic characteristics measured at each mooringstation at 10 m water depths and of MAT reconstructions of winter and summer SSTs and SSSs (confidence level 95 ) instrumentallymeasured SSTs and SSSs are denoted by the name of the mooring stations Roslashberg and Ytteroslashy estimated SSTs and SSSs including minand max values are marked by the core names MC 61-1 MC 99-1 and MC S4-1 Standard errors (Stde) for the mean values are given inbrackets

MC 61-1

Summer Roslashberg SST min SST max SST Roslashberg SSS min SSS max SSSMean (Std error) 110 (02) 124 (04) 81 (02) 174 (03) 311 (01) 281 (06) 210 (06) 328 (04)Range 10 48 28 37 95 63 74 32Minimum 59 95 68 140 241 243 204 316Maximum 159 143 96 177 336 306 278 347Winter Roslashberg SST min SST max SST Roslashberg SSS min SSS max SSSMean (Std error) 47 (01) 32 (03) minus01 (04) 66 (05) 325 (01) 304 (07) 257 (03) 335 (03)Range 51 41 30 34 46 43 39 23Minimum 26 06 minus16 48 295 278 254 325Maximum 77 47 15 82 341 321 292 348

MC 99-1

Summer Roslashberg SST min SST max SST Roslashberg SSS min SSS max SSSMean (Std error) 110 (02) 90 (08) 50 (08) 137 (10) 311 (01) 281 (04) 223 (06) 324 (04)Range 10 131 104 137 95 72 80 63Minimum 59 13 minus13 40 241 243 204 284Maximum 159 144 91 177 336 315 283 346Winter Roslashberg SST min SST max SST Roslashberg SSS min SSS max SSSMean (Std error) 47 (01) 11 (04) minus11 (02) 37 (07) 325 (01) 306 (03) 268 (05) 335 (02)Range 51 68 33 97 46 54 66 28Minimum 26 minus16 minus18 minus15 295 282 254 320Maximum 77 52 15 82 341 336 319 348

MC S4-1

Summer Ytteroslashy SST min SST max SST Ytteroslashy SSS min SSS max SSSMean (Std error) 104 (02) 110 (04) 68 (05) 162 (06) 306 (01) 283 (03) 208 (03) 324 (03)Range 85 87 121 86 87 72 92 66Minimum 62 62 03 91 247 244 204 284Maximum 147 149 124 177 334 315 296 350Winter Ytteroslashy SST min SST max SST Ytteroslashy SSS min SSS max SSSMean (Std error) 42 (01) 18 (02) minus11 (02) 51 (03) 320 (01) 306 (02) 257 (02) 333 (02)Range 49 56 33 67 57 41 62 32Minimum 21 minus03 minus18 15 282 288 254 319Maximum 7 53 15 82 34 328 316 351

the dinocyst blooming seasons but the ranges given betweentwo extrema (minimal maximal values within a year) is eco-logically meaningful regarding the phytoplankton dynamicsStill winter SSTs estimated by MAT are calculated with re-gard to the yearly lowermost values observed at the ana-logues core sites and may account for the generally large un-derestimation of winter SSTs (de Vernal et al 2000 2005)

Qualitatively most deviations from the instrumental datamay be explained by the poor species richness The dominat-ing speciesP daleiis assumed to thrive best in regions char-acterized by cold and low saline surface layers close to sea-ice marginsO centrocarpumis considered as a cosmopoli-tan species with a strong preference for warm and saline wa-

ters and a strong ability to adapt to rapid changes (eg Marretand Zonneveld 2003 Zonneveld et al 2013) The effect ofchanging abundances can be essentially observed in recon-structed SSTs in MC 99-1 before 1976 whenP daleidomi-nated the assemblages by up to 80 The reconstructed sum-mer and winter SSTs reveal very low values (mean around4C andminus2C in summer and winter respectively) from1960 to 1976 This is clearly at odds with the instrumentalmeasurements which show only minor changes in the mea-sured summer and winter temperatures around 10C and4C respectively Furthermore most species in the Trond-heimsfjord tolerate large temperature and salinity changes astypical for coastal and fjord environments and none of the



Fig 9Reconstructed primary productivity (g C mminus2 yrminus1) at everycore site (black lines) grey surrounding lines minimum and maxi-mum estimates for each reconstructed value calculated from the setof the 5 selected analogues

species follows the surface temperature and salinity variabil-ity measured in the fjord (Fig 2) Based on our previous ob-servations of changes in the cyst assemblages linked to theambient environmental parameters (see Sect 42) we assumethat cyst production in the Trondheimsfjord during the last50 yr is to a high extent controlled by the nutrient availabilitylinked to internal mixing and river discharges as well as dy-namically induced stress rather than empirically and solelySSTs and SSSs

The reconstructed mean PP values are lowest at MC 99-1 (260 g C mminus2 yrminus1) and highest at MC 61-1 (370 g C mminus2

yrminus1) (Table 4 Fig 9) The range between minimum andmaximum estimated PP for a given core location is sig-nificantly higher in the Seaward Basin and Middle Fjord(ca 250ndash400 g C mminus2 yrminus1) than at the fjord entrance (MC61-1 70 g C mminus2 yrminus1) In the absence of published data setson annual rates of primary productivity in the Trondheims-fjord a proper comparison of estimated vs measured PP val-ues is not possible Chlorophyll distributions extracted frommoderate resolution images of the MODIS spectroradiome-ter for the 2002ndash2005 period translate into an annual primaryproductivity of 200ndash300 g C mminus2 yrminus1 along the Norwegiancoast (httpdaacgsfcnasagov) This is twice the rate re-ported by Erga et al (2005) and Aure et al (2007) for Nor-wegian fjords and coastal waters but corresponds to mea-sured PP in local Scandinavian fjords ranging between 95and 241 g C mminus2 yrminus1 (Grundle et al 2009 and referencestherein) as well as to the 150ndash300 g C mminus2 yrminus1 PP range re-

Table 4 Descriptive statistics of MAT reconstructions of annualprimary productivity (confidence level of 95 ) standard errors forthe mean values (Stde) are given in brackets

MC 61-1 PP min max

Mean (Std error) 3684 (144) 2226 (70) 5633 (38)Range 1723 743 454Min 2710 1607 5216Max 4433 2351 5670MC 99-1 PP min maxMean (Std error) 2613 (228) 1488 (164) 4201 (347)Range 3929 2190 4239Min 553 161 1431Max 4481 2351 5670MC S4-1 PP min maxMean (Std error) 308 3 (120) 1842 (99) 4866 (212)Range 2606 2327 3320Min 1993 353 2351Max 4600 2680 5670

ported by the UNEP Regional Seas Programme for coastalwaters off Norway (Sherman and Hempel 2008) Further-more primary productivity in coastal areas collected byMODIS are considered to be slightly overestimated whichaccordingly leads to overestimations of the reconstructed PPdata (Durantou 2013 reviewer comment) However we can-not exclude effects of autocorrelation due to the dynamic linkof salinity and primary productivity in the Trondheimsfjord

Hence despite the relatively accurate reconstructed valuesfor winter and summer SSTs and SSSs as well as for meanannual PPs we suggest that estimated parameters have to becarefully evaluated with regard to the reference data sets usedfor the reconstructions the climatic and environmental fluc-tuations and associated changes in the cyst assemblages

5 Conclusions

Dinocyst assemblages from three sediment cores in theTrondheimsfjord show a high degree of coherency in termsof temporal succession Changes in cyst abundances andspecies diversity essentially reflect decadal-scale changes inNAO pattern and related variability in SSTs river dischargesSSSs and trophic conditions at the time of cyst productionSignificant impact by human activity on the fjord ecologicalstate over the last 50 yr is barely detectable from the investi-gated micropaleontological successions Deviations from theNAO-induced pattern of dinocyst assemblage compositionbetween each core site can be mainly explained by local hy-drological and topographical features

Species assemblages in core MC 61-1 at the fjord entrancereveal relatively high species evenness and cyst concentra-tions Downcore changes in species distribution are primarilydriven by regional-scale variations in the physico-chemical



characteristics of the NCC linked to the changing continen-tal run-off from southern Scandinavia

Core site MC 99-1 in the Seaward Basin is located in theflow path of strong tide and bottom water currents and highriver inputs from the local rivers Orkla and Gaula Both in-duce inward and seaward displacement of cysts and sedi-ments across the fjord as well as dilution of microfossilsby terrigeneous sediments This explains the lowest recordedmean average cyst concentrations However at this stage wecannot explain the poor species evenness and the dominanceof P dalei at this location compared with the inner fjordlocation (MC S4-1)

Cyst assemblages at the most inshore site MC S4-1 re-flect the relative influence of marine vs continental climateand associated changes in the physico-chemical surface wa-ter characteristics related to the prevailing NAO phase Thecore location is strongly influenced by the river VerdalselvaAssociated sediment and nutrient loads on one hand triggercyst production and on the other hand may cause dilution ofany fossil remains by terrigeneous material

Reconstructed mean and maximum abiotic and biotic sea-surface parameters are in relatively good agreement withmeasured time-series SSTs and SSSs in the Trondheimsfjordas well as with mean annual primary productivity rates fromcoastal Norway and other Scandinavian fjords Best esti-mates are generally achieved for summer SST and SSS Nev-ertheless the poor species evenness and the commonly largetolerance to SST and SSS changes in the dominant and sub-ordinate species may result in estimated values unlike the ob-served temperature and salinity trend We therefore suggestthat the estimated values of primary productivity reflect bestthe varying environmental conditions driving the productionof cysts Our pilot study also suggests that the use of MATtransfer functions for the reconstruction of sea-surface pa-rameters at a very high temporal scale such as implementedhere (annual to bi-annual over the last ca 50 yr) might behampered by the ldquomodernrdquo analogue reference hydrographicand dinocyst data sets which are often representative of en-vironmental conditions averaged over multi-annual to multi-centennial scales

Supplementary material related to this article isavailable online athttpwwwclim-pastnet103052014cp-10-305-2014-supplementpdf

AcknowledgementsThis work is a contribution to theCASE initial training network funded by the EuropeanCommunityrsquos 7th Framework Programme FP7 20072013Marie-Curie Actions under grant agreement no 238111(httpcaseitnepocubordeaux1frindexphphomehtml) Temper-ature and salinity data sets from the fjord mooring hydrologicalstations are provided by the Trondheim Biological Station of the

Norwegian University of Science and Technology We gratefullyacknowledge Bendik Eithun Halgunset from Soslashr-TroslashndelagFylkeskommune Viggo Finset from TroslashnderEnergi Kraft AS aswell as Arne Joslashrgen Kjoslashsnes and Eva Klausen from the NorwegianWater Resources and Energy Directorate for their cooperationhelpful suggestions and data access Anne de Vernal (GEOTOPUQAM) is gratefully acknowledged for her support helpfulsuggestions and reference data Ultimately we would like to thankR Telford (UIB Norway) for critical comments and helpful dis-cussions concerning statistical data treatment and transfer functions

Edited by D-D Rousseau

The publication of this article isfinanced by CNRS-INSU

References

Appleby P G Chronostratigraphic techniques in recent sedimentsin Tracking environmental change using lake sediments Volume1 Basin analysis coring and chronological techniques editedby Last W M and Smol J P p 2001 Kluwer Academic Pub-lishers Dordrecht the Netherlands 2001

Aure J Strand Oslash Erga S and Strohmeier T Primaryproduction enhancement by artificial upwelling in a west-ern Norwegian fjord Mar Ecol Prog Ser 352 39ndash52doi103354meps07139 2007

Bayon G Henderson G M and Bohn M U-Th stratigraphy ofa cold seep carbonate crust Chem Geol 2 47ndash56 2009

Bigler C and Wunder J Statistische Datenanalyse mit REine Einfuumlhrung fuumlr Umweltwissenschaftler Gebirgswaldoumlkolo-gie (D-FOWI) ETH Zuumlrich Version 10 1ndash39 2003

Birks H J B Quantitative palaeoenvironmental reconstructionsStat Model Quat Sci Data Tech Guid 5 edited by MaddyD and Brew J S Quat Res Assoc Cambridge 5 161ndash2541995

Blindheim J Borovkov V Hansen B Malmberg S-A TurrellW R and Oslashsterhus S Upper layer cooling and freshening inthe Norwegian Sea in relation to atmospheric forcing Deep SeaRes Part I 47 655ndash680 doi101016S0967-0637(99)00070-92000

Blume H P Bruumlmmer G Horn R Kretzschmar R KandelerE Koumlgel-Knabner I Stahr K and Wilke B M Schef-ferSchachtschabel Lehrbuch der Bodenkunde Springer Spek-trum 2009

Boslashe R Rise L Blikra L H Longva O and Eide A Holocenemass-movement processes in Trondheimsfjorden Central Nor-way Norw J Geol 83 3ndash22 2003

Dale B Cyst formation sedimentation and preservation factorsaffecting dinoflagellate assemblages in recent sediments fromTrondheimsfjord Norway Rev Palaeobot Palynol 22 39ndash601976



Dale B Marine dinoflagellate cysts as indicators of eutrophicationand industrial pollution a discussion Sci Total Environ 264235ndash240 2001

Dale B Thorsen T A and Fjellsa A Dinoflagellate cysts as in-dicators of cultural eutrophication in the Oslofjord Norway Es-tuar Coast Shelf S 48 371ndash382 doi101006ecss199904271999a

Dale T Rey F and Heimdal B R Seasonal development of phy-toplankton at a high latitude oceanic site Sarsia 84 419ndash4351999b

Dale B Dale A L and Jansen J H F Dinoflagellate cysts asenvironmental indicators in surface sediments from the Congodeep-sea fan and adjacent regions Palaeogeogr Palaeocl 185309ndash338 2002

De Vernal A and Marret F Organic-Walled Dinoflagellate CystsTracers of Sea-Surface Conditions Mar Geol 1 371ndash408doi101016S1572-5480(07)01014-7 2007

De Vernal A Rochon A Hillaire-Marcel C Turon J L andGuiot J Quantitative reconstruction of sea-surface conditionsseasonal extent of sea-ice cover and meltwater discharges inhigh latitude marine environments from dinoflagellate cyst as-semblages Nato Asi Ser I12 611ndash621 1993

De Vernal A Henry M and Bilodeau G Techniques de Preacutepara-tion et drsquoAnalyse en Micropaleontologie Le Cah du GEOTOP3 Universiteacute du Queacutebec agrave Montreacuteal Canada 1996

De Vernal A Hillaire-Marcel C Turon J-L and MatthiessenJ Reconstruction of sea-surface temperature salinity and sea-ice cover in the northern North Atlantic during the last glacialmaximum based on dinocyst assemblages Can J Earth Sci 37725ndash750 doi101139e99-091 2000

De Vernal A Henry M Matthiessen J Mudie P J RochonA Boessenkool K P Eynaud F Groslashsfjeld K Guiot JHamel D Harland R Head M J Kunz-Pirrung M LevacE Loucheur V Peyron O Pospelova V Radi T Turon J-L and Voronina E Dinoflagellate cyst assemblages as tracersof sea-surface conditions in the northern North Atlantic Arcticand sub-Arctic seas the new n = 677 data base and its applica-tion for quantitative palaeoceanographic reconstruction J Qua-ternary S 16 681ndash698 doi101002jqs659 2001

De Vernal A Eynaud F Henry M Hillaire-Marcel C Lon-deix L Mangin S Matthiessen J Marret F Radi T Ro-chon A Solignac S and Turon J-L Reconstruction of sea-surface conditions at middle to high latitudes of the NorthernHemisphere during the Last Glacial Maximum (LGM) based ondinoflagellate cyst assemblages Quaternary Sci Rev 24 897ndash924 doi101016jquascirev200406014 2005

Devillers R and de Vernal A Distribution of dinoflagellate cystsin surface sediments of the northern North Atlantic in relation tonutrient content and productivity in surface waters Mar Geol166 103ndash124 doi101016S0025-3227(00)00007-4 2000

Durantou L Rochon A Ledu D Masseacute G Schmidt S andBabin M Quantitative reconstruction of sea-surface conditionsover the last 150 yr in the Beaufort Sea based on dinoflagellatecyst assemblages the role of large-scale atmospheric circulationpatterns Biogeosciences 9 5391ndash5406 doi105194bg-9-5391-2012 2012

Ellingsen I H Internal tides and the spread of river plumes inthe Trondheimfjord 1ndash174 pp Norwegian University of Scienceand Technology (NTNU) Trondheim 2004

Erga S R Aursland K Frette Oslash Hamre B Lotsberg JK Stamnes J J Aure J Rey F and Stamnes K UVtransmission in Norwegian marine waters controlling factorsand possible effects on primary production and vertical distri-bution of phytoplankton Mar Ecol Prog Ser 305 79ndash100doi103354meps305079 2005

Groslashsfjeld K and Harland R Distribution of modern dinoflagellatecysts from inshore areas along the coast of southern Norway JQuaternary Sci 16 651ndash659 doi101002jqs653 2001

Groslashsfjeld K Harland R and Howe J Dinoflagellate cyst as-semblages inshore and offshore Svalbard reflecting their modernhydrography and climate Norw J Geol 89 121ndash134 2009

Grundle D S Timothy D A and Varela D E Variations ofphytoplankton productivity and biomass over an annual cycle inSaanich Inlet a British Columbia fjord Cont Shelf Res 292257ndash2269 doi101016jcsr200908013 2009

Guiot J and de Vernal A Transfer Functions Methods forQuantitative Paleoceanography Based on Microfossils Hillaire-Marcel C Vernal A Proxies Late Cenozoic Paleoceanogr El-sevier 1 523ndash563 doi101016S1572-5480(07)01018-4 2007

Hammer Oslash Harper D A T and Ryan P D PaleontologicalStatistics Software Paleontol Electron 4 0ndash9 available athttpwwwfolkuionoohammerpast(last access April 2012)Version 215 2001

Harland R Distribution maps of recent dinoflagellate cysts in bot-tom sediments from the North Atlantic Ocean and adjacent seasPaleontology 26 321ndash387 1983

Harland R Nordberg K and Filipsson H L A high-resolutiondinoflagellate cyst record from latest Holocene sediments inKoljouml Fjord Sweden Rev Palaeobot Palynol 128 119ndash141doi101016S0034-6667(03)00116-7 2004

Harland R Nordberg K and Filipsson H L Dinoflag-ellate cysts and hydrographical change in Gullmar Fjordwest coast of Sweden Sci Total Environ 355 204ndash231doi101016jscitotenv200502030 2006

Head M J Harland R and Matthiessen J Cold marine indica-tors of the late Quaternary The new dinoflagellate cyst genus Is-landinium and related morphotypes J Quaternary Sci 16 621ndash636 doi101002jqs657 2001

Head M J Lewis J and de Vernal A The cyst of the calcareousdinoflagellate Scrippsiella trifida resolving the fossil record ofits organic wall with that of Alexandrium tamarense J Paleon-tol 80 1ndash18 2006

Howe J A Austin W E N Forwick M Paetzel M HarlandR and Cage A G Fjord systems and archives a review inFjord Systems and Archives edited by Howe J A Austin WE N Forwick M and Paetzel M Geological Society LondonSpecial Publications 344 207ndash223 2010

Hurrell J W Decadal trends in the North Atlantic Oscillation Re-gional Temperatures and precipitation Science 269 676ndash6791995



Hurrell J W and van Loon H Decadal variations in climateassociated with the North Atlantic Oscillation in ClimaticChange 36 301ndash326 Kluwer Academic Publishers Dordrechtthe Netherlands 1997

Imbrie J and Kipp N G A new micropaleontological methodfor quantitative paleoclimatology Application to a late Pleis-tocene Caribbean core edited by Turekian K K Late CenozoicGlacial Ages 71ndash181 Yale Univ Press New Haven CT 1971

Jacobson P Physical oceanography of the Trondheimsfjord Geo-phys Astro Fluid 26 3ndash26 1983

Kirchner G210Pb as a tool for establishing sediment chronolo-gies examples of potentials and limitations of conven-tional dating models J Environ Radioact 102 490ndash494doi101016jjenvrad201011010 2011

Kucera M Weinelt M S Kiefer T Pflaumann U HayesTChen M-T Mix A C Barrows T T Cortijo E Duprat JJuggins S and Waelbroeck C Reconstruction of sea-surfacetemperatures from assemblages of planktonic foraminiferaMulti-technique approach based on geographically constrainedcalibration data sets and its application to glacial Atlantic andPacific Oceans Quaternary Sci Rev 24 951ndash998 2005

LrsquoAbeacutee-Lund J H Eie J A Faugli P E Haugland S HvidstenN A Jensen A J Melvold K Pettersen V and Saltveit S JRivers in Boreal Uplands in Rivers of Europe edited by Tock-ner K Robinson C T and Uehlinger U 0ndash699 AcademicPress Elsevier 2009

Marret F and Zonneveld K Atlas of modern organic-walled di-noflagellate cyst distribution Rev Palaeobot Palynol 125 1ndash200 doi101016S0034-6667(02)00229-4 2003

Marret F Eirigraveksson J Knudsen K L Turon J-L and ScourseJ D Distribution of dinoflagellate cyst assemblages in sur-face sediments from the northern and western shelf of Ice-land Rev Palaeobot Palynol 128 35ndash53 doi101016S0034-6667(03)00111-8 2004

Matsuoka K Kawami H Nagai S Iwataki M and TakayamaH Re-examination of cyst-motile relationships ofPolykrikoskofoidii Chatton andPolykrikos schwartziiBuumltschli (Gymno-diniales Dinophyceae) Rev Palaeobot Palynol 154 79ndash902009

Matthiessen J Distribution patterns of dinoflagellate cystsand other organic-walled microfossils in recent Norwegian-Greenland Sea sediments Mar Micropaleontol 24 307ndash334doi1010160377-8398(94)00016-G 1995

Maslin M A Shackleton N J and Pflaumann U Surface watertemperature salinity and density changes in the northeast At-lantic during the last 45000 years Heinrich events deep waterformation and climatic rebounds Paleoceanography 10 527ndash544 doi10102994PA03040 1995

McMinn A Recent dinoflagellate cysts from estuaries on the cen-tral coast of New South Wales Australia Micropaleontology 37269ndash287 1991

Milzer G Giraudeau J Faust J Knies J Eynau F and Ruumlh-lemann C Spatial distribution of benthic foraminiferal sta-ble isotopes and dinocyst assemblages in surface sediments ofthe Trondheimsfjord central Norway Biogeosciences 10 1ndash16doi105194bg-10-1-2013 2013

Nehring S Dinoflagellate resting cysts in recent sedimentsof tide western Baltic as indicators for the occurrence ofldquonon-indigenousrdquo species in the water column Institut fuumlrMeereskunde Universitaumlt Kiel in Proceedings of the 13th Sym-posium of the Saldo Marine Biologists Riga 1993 79ndash85 1997Ed A AndrugaitaL Institue of Aquatic Ecology University ofLatvia ISBN 9984-509-90-7 1997

Ottersen G Planque B Belgrano A Post E Reid P andStenseth N Ecological effects of the North Atlantic OscillationOecologia 128 1ndash14 doi101007s004420100655 2001

Parke M and Dixon P S Revised check-list of British marinealgae J Mar Biol Assoc UK 44 499ndash542 1964

Penaud A Eynaud F Turon J L Zaragosi S Mar-ret F and Bourillet J F Interglacial variability (MIS 5and MIS 7) and dinoflagellate cyst assemblages in the Bayof Biscay (North Atlantic) Mar Micropal 68 136ndash155doi101016jmarmicro200801007 2008

Pettersson L-E Totalavloslashpet fra Norges vassdrag 1900-2010(Rapport nr 39-2012) Oslo ISBN 978-82-410-0827-6 2012

Pflaumann U Duprat C Pujol C and Labeyrie L SIMMAX Amodern analog technique to deduce Atlantic sea surface temper-atures from planktonic foraminifera in deep-sea sediments Pale-oceanography 11 15ndash35 1996

Pospelova V de Vernal A and Pedersen T F Distribution ofdinoflagellate cysts in surface sediments from the northeasternPacific Ocean (43ndash25 N) in relation to sea-surface temperaturesalinity productivity and coastal upwelling Mar Micropaleon-tol 68 21ndash48 doi101016jmarmicro200801008 2008

R Development Core Team R A language and environment forstatistical computing R Foundati Vienna Austria available athttpwwwr-projectorg 2008

Radi T and de Vernal A Dinocyst distribution in surface sed-iments from the northeastern Pacific margin (40ndash60 N) in re-lation to hydrographic conditions productivity and upwellingRev Palaeobot Palynol 128 169ndash193 doi101016S0034-6667(03)00118-0 2004

Radi T and de Vernal A Dinocysts as proxy of primary productiv-ity in midndashhigh latitudes of the Northern Hemisphere Mar Mi-cropaleontol 68 84ndash114 doi101016jmarmicro2008010122008

Riding J B and Kyffin-Hughes J E A review of the laboratorypreparation of palynomorphs with a description of an effectivenon-acid technique Rev Bras Paleontol 7 13ndash44 2004

Robbins J A and Edgington D N Determination of recent sed-imentation rates in Lake Michigan using Pb-210 ad CS-137Geochim Cosmochim Ac 39 285ndash304 1975

Rochon A de Vernal A Turon J-L Matthiessen J and HeadM J Distribution of recent dinoflagellate cysts in surface sedi-ments from the North Atlantic Ocean and adjacent seas in rela-tion to sea-surface parameters Am Assoc Stratigr Palynol 351ndash146 doi1010160377-8398(94)00016-G 1999

Saeligtre M M L Dale B Abdullahb M I and Saeligtre G P Di-noflagellate cysts as potential indicators of industrial pollution ina Norwegian fjord Mar Environ Res 44 167ndash189 1997

Saeligtre R Features of the central Norwegian shelf circulation ContShelf Res 19 1809ndash1831 1999

Saeligtre R Norwegian Coastal Current edited by R Saeligtre TapirAcademic Press 2007



Sakshaug E and Myklestad S Studies on the phytoplanktonecology of the TrondheimsfjordIII Dynamics of phytoplanktonblooms in relation to environmental factors bioassy experimentsand parameters for the physiological state of the populations JExp Mar Biol Ecol 11 157ndash188 1973

Schmidt S Howa H Diallo A Cremer M Duros P FontanierC Deflandre B Metzger E and Mulder T Recent sed-iment transport and deposition in the Cap-Ferret CanyonSouth-East margin of Bay of Biscay Deep-Sea Res Part IIdoi101016jdsr2201306004 2013

Sherman K and Hempel G The UNEP Large Marine Ecosys-tem Report A perspective on changing conditions in LMEs ofthe worldrsquos Regional Seas UNEP LME Report No 182 0ndash8722008

Smayda T J and Reynolds C S Strategies of marine dinoflagel-late survival and some rules of assembly J Sea Res 49 95ndash106doi101016S1385-1101(02)00219-8 2003

Sorrel P Popescu S-M Head M J Suc J P Klotz Sand Oberhaumlnsli H Hydrographic development of the Aral Seaduring the last 2000 years based on a quantitative analysisof dinoflagellate cysts Palaeogeogr Palaeocl 234 304ndash327doi101016jpalaeo200510012 2006

Stockmarr J Tablets with spores used in absolute pollen analysisPollen et Spores 8 615ndash621 1971

Stroslashmgren T Zooplankton and Hydrography in Trondheimsfjor-den on the west coast of Norway R Nor Soc Sci Lett Museum51-60 1ndash38 doi1010160024-3841(84)90026-3 1974

Tangen K and Arff J Hoslashvringen wastewater plant and the en-vironmental quality of the Trondheimfjord (OCN R-23027)Trondheim (Norway) 2003

Telford R J and Birks H J B Effect of uneven sampling along anenvironmental gradient on transfer-function performance J Pa-leolimnol 46 99ndash106 doi101007s10933-011-9523-z 2011

Van Nieuwenhove N Bauch H and Matthiessen JLast interglacial surface water conditions in the east-ern Nordic Seas inferred from dinocyst and foraminiferalassemblages Mar Micropaleontol 66 247ndash263doi101016jmarmicro200710004 2008

Vellinga M and Wood R Global climatic impacts of a collapseof the Atlantic thermohaline circulation Clim Change 54 251ndash267 2002

Visbeck M Chassignet E P Curry R Dickson B and Krah-mann G The Oceanrsquos Response to North Atlantic OscillationVariability in The North Atlantic Oscillation Climatic Signifi-cance and Environmental Impact edited by Hurrell J W 113ndash45 Washington DC 2003

Von Detten P Faude O and Meyer T Leitfaden zur statistischenAuswertung von empirischen Studien 1ndash44 2008

Waelbroeck C Mulitza S Spero H Dokken T Kiefer Tand Cortijo E A global compilation of late Holocene plank-tonic foraminiferal δ18O relationship between surface watertemperature andδ18O Quaternary Sci Rev 24 853ndash868doi101016jquascirev200310014 2005

Wall D and Dale B The resting cysts of modern marine dinoflag-ellates and their paleontological significance Rev Palaeobot Pa-lynol 2 349ndash354 1967

Zaragosi S Eynaud F Pujol C Auffret G A Turon J-L andGarlan T Initiation of the European deglaciation as recordedin the northwestern Bay of Biscay slope environments (Meri-adzek Terrace and Trevelyan Escarpment) a multi-proxy ap-proach Earth Planet Sci Lett 188 493ndash507 2001

Zonneveld K A F Versteegh G J M and de Lange G JPreservation of organic-walled dinoflagellate cysts in differentoxygen regimes a 10000 year natural experiment Mar Mi-cropaleontol 29 393ndash405 1997

Zonneveld K A F Versteegh G J M and de Lange G JPalaeoproductivity and post-depositional aerobic organic matterdecay reflected by dinoflagellate cyst assemblages of the EasternMediterranean S1 sapropel ABC26 Mar Geol 172 181ndash1952001

Zonneveld K A F Versteegh G and Kodrans-Nsiah M Preser-vation and organic chemistry of Late Cenozoic organic-walleddinoflagellate cysts A review Mar Micropaleontol 68 179ndash197 doi101016jmarmicro200801015 2008

Zonneveld K A F Marret F Versteegh G J M BogusK Bonnet S Bouimetarhan I Crouch E de Vernal AElshanawany R Edwards L Esper O Forke S Groslashs-fjeld K Henry M Holzwarth U Kielt J-F Kim S-Y Ladouceur S Ledu D Chen L Limoges A Lon-deix L Lu S-H Mahmoud M S Marino G MatsoukaK Matthiessen J Mildenhal D C Mudie P Neil H LPospelova V Qi Y Radi T Richerol T Rochon A San-giorgi F Solignac S Turon J-L Verleye T Wang Y WangZ and Young M Atlas of modern dinoflagellate cyst distribu-tion based on 2405 data points Rev Palaeobot Palyno 191 1ndash197 doi101016jrevpalbo201208003 2013



variability and on the interaction of marine and continentalclimate conditions

The distribution of organic-walled dinoflagellate cysts (ordinocysts) the zygotic products of dinoflagellates in sed-iments is controlled by the surface water characteristicsmaking them widely used as paleoceanographic proxies forthe reconstruction of sea-surface parameters (eg Harland1983 de Vernal et al 1993 Rochon et al 1999 Marretand Zonneveld 2003 Zonneveld et al 2013) Dinocystsare composed of highly resistant refractory organic matterand are therefore usually well preserved in sediments (egRochon et al 1999 de Vernal et al 2000 Zonneveld etal 2007) Dinoflagellates in coastal environments are gen-erally adapted to large sea-surface temperature (SST) andsalinity (SSS) ranges due to the seasonally varying relativeinfluence of marinecoastal and continental waters In thesesettings their occurrence is therefore rather related to otherspecies-discriminating environmental parameters than in theopen ocean such as lateral and vertical salinity gradientsnutrient availability and the species-specific ability to adaptto rapidly changing hydrological conditions (eg Smaydaand Reynolds 2003 Harland et al 2004 2006 Radi et al2007 Pospelova et al 2010) These hydrological changesmay however be biased by local and regional anthropogeni-cally induced changes in the coastal and fjord environments(Saeligtre et al 1997 Dale 2001) Recent studies on surfacesediments from the Trondheimsfjord which encompass thepast 4ndash20 yr show large spatial variability in the compositionof cyst assemblages linked to the complex fjord hydrographyand topography (Milzer et al 2013) Untreated waste waterfrom local and regional industry and agriculture was directlyled into the Trondheimsfjord until 1980 The assemblageshowever do not reveal any indications of human-induced eu-trophication suggesting that recently imposed environmen-tal regulations and restrictions to reduce these nutrient inputswere highly effective (Tangen and Arff 2003) Alternativelythe exchange of Trondheimsfjord waters with the open seastrongly dilutes excess nutrient loads and the regular supplyof oxygen-rich Atlantic waters prevents hypoxic or anoxicconditions at the bottom (Tangen and Arff 2003)

In the following study we investigate recent changes inthe dinocyst assemblages in the Trondheimsfjord over thelast 25 to 50 yr from three sediment cores located along thefjord axis We discuss the downcore distribution of dinocystsin view of naturally and anthropogenically induced changeswith regard to SST and SSS variability (monitored at threefixed mooring stations in the fjord) annual mean river dis-charges atmospheric oscillation patterns (state of the NorthAtlantic Oscillation ndash NAO) and nutrient availability whichall together account for changes in the fjord ecological state(Sakshaug and Myklestad 1973 Ottersen et al 2001) Weultimately apply the now standard modern analogue tech-nique (MAT) to the downcore distribution of dinocysts andtest its ability to reconstruct quantitatively a series of sea-surface parameters (SST SSS and primary productivity (PP))

in this fjord setting in order to establish a solid basis forfuture investigations of paleoclimate and paleoceanographicvariability

2 Physical settings and modern hydrology

The Trondheimsfjord extends from 6340prime N0945prime E at Oslashr-land at the fjord entrance to 6445prime N1130prime E at the fjordhead at Verdal (Boslashe et al 2003) and runs over into the Beit-stadfjord from Skarnsund 6350prime N1104prime E until Steinkjerat 6300prime N1128prime E (Fig 1) The fjord is 140 km long withan overall volume of 235 km3 and a surface area of 1420 km2The entire fjord area is divided into three basins the SeawardBasin the Middle Fjord and the Beitstadfjord The basins areseparated by three sills located at the fjord entrance (195 mwater depth) at the island of Tautra (98 m water depth) andat Skarnsund (140 m water depth) (Fig 1) (Jacobson 1983Ellingsen 2004) Sediment deposition in the Trondheims-fjord is largely controlled by the prevalent currents and theassociated pelagicmarine sediment load the topography andrivers delivering large amounts of terrigenous sediment intothe fjord (Boslashe et al 2003 Faust et al unpublished dataMilzer et al 2013) Bottom currents and tidal currents inthe subsurface layers prevent the rapid deposition of the fineparticle fraction in topographically elevated areas like sillsInstead the fine particle fraction primarily accumulates inthe more sheltered fjord basins Sedimentation rates are thushigher in the Middle Fjord and Beitstadfjord than at the fjordentrance and in the Seaward Basin where several currentspass at high frequencies (Fig 1) (Boslashe et al 2003 Milzeret al 2013) Silt to sand microfossils including dinocystand foraminifera accordingly mirror a local averaged patternwith however clear distribution gradients from the entranceto the inner part of the fjord (Milzer et al 2013)

21 Water mass exchange with the Norwegian Sea

The hydrology of the Trondheimsfjord is characterized bya three-layer system mainly driven by density differencesrelated to the seasonal temperature and salinity variabilityof the water masses inside and outside the fjord (eg Sak-shaug and Myklestad 1973 Stroslashmgren 1974 Jacobson1983) The bottom water is generally renewed twice a yearin FebruaryndashJune and SeptemberndashNovember by NAC andNCC waters respectively (Sakshaug and Myklestad 1973Stroslashmgren 1974 Jacobson 1983) The depth of the NACand its ability to cross the sill at the Trondheimsfjord en-trance depends on the depth of the overlying wedge-shapedNCC which in turn is linked to the prevailing seasonal windpatterns in the Nordic Seas (Jacobson 1983 Saeligtre 19992007) In early winter-spring when the NCC is supplied bylarge amounts of fresh water from the Norwegian hinterlandthe current is deflected offshore by northerly winds form-ing a shallow surface layer NAC waters replace the NCC
















































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References










































































































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References


































































































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References





















































































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References









































































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References

































































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References






















































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References










































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References



































































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References



























































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References

















































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References






































































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References






























































































MC 61-1


MC 99-1


MC S4-1












MC 61-1 PP min max




5 Conclusions














References


































































































MC 61-1 PP min max




5 Conclusions














References






































































































References
























































































































































































































































Documents

Climate Qualitative and quantitative reconstructions of surface … · 2020. 6. 19. · culation of the topographically steered North Atlantic Cur-rent (NAC) and the Norwegian Coastal