Social Media

Matching diatom assemblages in lake sediment cores and modern surface sediment samples: the implications for lake conservation and restoration with special reference to acidified systems

Description
Matching diatom assemblages in lake sediment cores and modern surface sediment samples: the implications for lake conservation and restoration with special reference to acidified systems
Categories
Published
of 14
0
Categories
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Share
Transcript
   Hydrobiologia  344:  27–40, 1997.  27 c    1997  Kluwer Academic Publishers. Printed in Belgium. Matching diatom assemblages in lake sediment cores and modern surfacesediment samples: the implications for lake conservation and restorationwith special reference to acidified systems R. J. Flower, S. Juggins 1 & R. W. Battarbee  Environmental Change Research Centre, University College London, 26 Bedford Way, London WC1H 0AP, UK  1  Department of Geography, University of Newcastle, Newcastle Upon Tyne, NE1 7RU, UK  Received 4 April 1996; in revised form 5 November 1996; accepted 10 December 1996 Abstract Restoration goals for damaged freshwater habitats can be defined according to ecological as well as to chemicalcriteria. For disturbed lakes, the sediment microfossil record can be used to select potential modern analogue sitesas possible restoration target ecosystems.Fossildiatomassemblagesintwoacidifiedlakes(RoundLochofGlenheadandLochDee)inGalloway,Scotland,were compared floristically with modern surface sediment samples from ca. 200 lakes in Britain, Ireland, Swedenand Norway using numerical techniques. Mean squared Chi-squared dissimilarity (SCD) values based on betweensample Chi-square distance measures were used to compare samples.‘Space-for-time substitution’ using diatom assemblage matching techniques identified several modern analoguesites with Hebridean Loch Teanga and Irish Lough Claggan possessing modern diatom floras most similar to thosewhich existed at the Round Loch of Glenhead and Loch Dee  before  acidification.From the point of view of atmospheric pollution, the most closely matching modern analogue sites were notnecessarilyinthemostpristineregions.SomeanaloguesoccurredinUK regionsofmoderateorlowaciddepositionand modern diatom assemblages in atmospherically cleaner mid Norway were generally less similar floristically.It is argued that identification of modern analogue sites raises the possibility of using time-space substitution of closely matched modern and fossil samples to infer whole lake ecosystems.Diatoms are however poor indicators of some water chemistry variables and the two closest matched modernanalogue sites have too high calcium concentrations making faunistic comparisons questionable.Identification of good modern analogue lakes can be improved by using selection criteria, other than diatoms,to pre-select sites. Screening inappropriate sites according to water chemistry and basin features combined with alarger biological database of modern and fossil samples offers a promising way of refining the selection processes.Despite necessary refinements, modern analogue matching can potentially identify whole lake ecosystems thatcan serve as biological target communities for currently disturbed sites. Being based on biological rather thanchemical criteria, this approach does not rely on species-water chemistry transfer functions. It is therefore directlyrelevant to lake conservation and restoration objectives and offers an alternative method for reconstructing lakepalaeo-environments. Introduction The principles of restoration ecology are now fair-ly well established (e.g. Jordan et al., 1987; Gunn,1995) and have been applied to a variety of environ-mental problems including rehabilitation of derelictindustrial land (Bradshaw & Chadwick, 1980), reaf-forestation of cleared landscapes (Shiva et al., 1985),andcurbingnutrientenrichmentoflakes(Edmondson,1977; Vollenweider,1987)andrivers(Wheeler, 1974).Despite these apparent successes, there are consider-able differences in the literature concerning restora-tion objectives (e.g. Kelly & Harwell, 1990). Mere-ly establishing vegetation on derelict land (Bradshaw,1983) or redeveloping a site according to local inter-est and past use (Bj¨ork, 1988) does not restore natural *129560    28ecosystems. Similarly, lake management to improvechemical(Davidson,1987)orphysical(Harvey& Jos-selyn, 1986) conditions, to restore past productivitylevels (Broberg, 1988) or to alter predator-prey rela-tionships (Shapiro & Wright, 1984) will not neces-sarily re-establish former natural communities. Mostmanipulationsnevertheless seem to result in improvedenvironmentalquality accordingto certain criteria but,if conservation objectives are important, restoration of natural communities must be the primary objective.Lakeacidificationbyindustrialpollutionisa majorconservation issue in Europe with, for example, morethan 68000 km 2 of lake-lands being affected in Nor-way alone (Henriksen et al., 1992). Before any sitespecific mitigation management can be implementedeffectively it is necessary to define ecological as wellaschemicaltargetsforrestorationmeasures.Thesetar-gets must be clearly defined; for acidified lakes thereareatleastthreewaystodothis. Firstly, usingpalaeoe-cology to provide a record of those organisms thatexisted before site acidification. The sediment recordhas provided invaluable information about the timingand extent of lake acidification (e.g. Battarbee et al.,1989) but records are restricted to groups that pre-serve well in sediments, diatoms, chironomids andsome zooplankton (see Berglund, 1986). Secondly,‘space-for-time substitution’ (Pickett, 1988) assumesthat aquatic communities currently existing in unpol-luted or ‘clean’ areas resemble closely those that for-merly existed in otherwise similar but now acidifiedhabitats (Muniz, 1987). The ‘clean area’ communitiescan, however, vary considerably due to unknown orpoorly understood biogeographic and chemical influ-ences. These effects make specifying particular target‘clean’ lake ecosystems difficult. Thirdly, experimen-tal manipulation of ecosystems, usually through acidorbaseadditions,canrevealcommunityresponsesthatcan be used to infer large scale changes in the ‘natu-ral’ environment (e.g. Schindler, 1980). Very detailedinformationaboutcommunitychangescanbeobtainedinthiswaybuttheapproachsuffersfromrepresentativ-ity and replicability problems, since results are basedon a few, relatively small-scale, manipulations. Allthree approaches can offer some guidance for ecosys-tem restorationbut at present nonecan be used to inferconfidently the pre-disturbance biota of an acidifiedlake.Atleastapartialsolutiontotheproblemofdefiningtarget ecosystems for restoration purposes is to iden-tify modern analogue sites for a currently disturbedlake using ecological criteria. This can be achievedby matching fossil assemblages present in sedimentsdeposited before acidification with those assemblagesfound in modern surface sediments of pristine or lessaffected lakes. For matching purposes, diatoms areprobablythesinglemostappropriatemicrofossilgroupsince their living communities are diverse, speciesare sensitive to water chemistry and habitat type, anddiatom assemblages preserve well in lake sediments.They occupy both benthic and planktonic environ-ments and are often the main food resource of inver-tebrates, especially in upland freshwaters. We arguetherefore that if pre-acidification sedimentary diatomassemblagescanbeusedtoidentifyamodernanaloguesite, thenitislikelythatmanyotherecosystemcompo-nents, such as invertebrates and possibly fish species,are also common to the two environments. Althoughthis is a major assumption, the importance of identi-fying possible modern analogue sites makes the issueworth exploring.The pre-acidification diatom floras of many acid-ified lakes in Northwest Europe and North Americaare now relatively well known from a variety of palae-olimnologicalinvestigations. Modern data-sets of lakesurfacesedimentdiatomassemblagesandwaterchem-istry are also available for over 400 lakes from theseregions (Davis & Anderson, 1985; Charles, 1985;Flower, 1986a; Stevenson et al., 1991). This studyuses numericaltechniquesto examinefloristic similar-itiesbetweendiatomassemblagesinarangeofmodernnorthwest European lake surface sediments and thosepreserved in sediment cores from two acidified lakesin southwest Scotland. The focus of the approach is toidentify modern analogue communities for these twolakes which can then serve as biologically defined tar-get ecosystems for acidity mitigation measures. Methods  Modern and fossil data sets The surface sediment diatom assemblages and waterchemistry data were obtained from 170 lakes in theUK, NorwayandSwedenaspart oftheSurfaceWatersAcidification Project (SWAP) (Battarbee & Renberg,1990; Stevenson et al., 1991). These data were aug-mented with 24 sites from the NW Ireland and themid-Norway (Figure 1). Annual non-marine sulphatedeposition data are indicated in the figure and showthatsampleareasinNWScotland,westernIreland,andmid-Norway are located in relatively low acid deposi-  29tion areas. All the sites sampled are natural softwaterlakes varying in pH from 4.4 to 7.8. The less acidlakes comprise two groups, those which are very sen-sitive to acidification(lowAcid NeutralizingCapacity,ANC)but occurin areasoflow aciddepositionsuchasthe northwest of Scotland or mid-Norway, and thosewhich are less sensitive to acidification (higher ANC)but occur in regionswhere acid depositionis moderateor high.Fossil diatom samples (from 0.5 or 1 cm sedi-ment core sections) are the same as those used in pub-lishedbiostratigraphicprofilesfromtheRoundLochof Glenhead (RLGH) and Loch Dee (LDEE), Galloway,SW Scotland (see Flower et al., 1987). Both lakesare in an area of high acid deposition and have beenacidified since the mid to late 19th century. These twosites were selected for analysis because, althoughbothare acidified, LDEE naturally possesses considerablymore ANC and, compared with the RLGH, was ini-tially less acid and is currently less acidified. Hence,LDEE and RLGH have responded differently to aciddepositionand their fossil assemblagesprovidea wideand representativespectrum of soft-water diatom taxa.All the modern and fossil data resulting fromthe surface sediment and sediment core analysesare held on the Environmental Change ResearchCentre (ECRC) database ‘AMPHORA’ maintainedat University College London (Munro et al., 1990and see the ECRC’s INTERNET WEB Page,http://www.geog.ucl.ac.uk/     abeare/Amphora.html).  Numerical analysis Floristic differences between the fossil and moderndiatom assemblages samples were measured using thesquared Chi-square distance measure: d  2 ij  =  m  X  k  =  1   y  ik    y  jk    2 =    y  ik  +  y  jk   ; where  y ik   is the proportion of diatom taxon  k   in sam-ple i, d  ij   istheChi-squareddistancebetweensamples i and  j . The squared Chi-squared distance has the desir-ablepropertyofweightingthediatomproportionssoasto emphasize the signal component of the differencesbetween assemblages at the expense of the noise com-ponent (Overpeck et al., 1985). The values of squaredChi-squareddistancecanvarybetweenzeroand2,withlower values indicating more similar assemblages.The fossil diatom samples from each core weregrouped into zones of similar floristic compositionusing constrained cluster analysis (Birks & Gordon,1985).The squared Chi-squared distance values were cal-culated between fossil samples and each sample inthe modern dataset. Distances between fossil and eachmodern sample were then averaged for each core-diatom zone, to give a single value that representsthe squared Chi-squared dissimilarity (SCD) betweeneach core zone and each modern sample.Critical values of SCD were defined by taking thefirst and second percentiles of the matrix of squaredChi-squareddistancesamongall modernsamples. Theresulting thresholds of 0.57 and 0.65 are used to indi-cate ‘very good’ and ‘good’ analogue sites, respec-tively (cf. Bartlein & Whitlock, 1993). These comparewith a value of approximately 0.25 for replicate coretops from the RLGH or the LDEE sites. Simple statis-tical tests (Manly, 1991) show that these dissimilarityvaluesare highlysignificant, being verydifferentfromrandom allocations of abundances to taxa.Diatom-inferred pH values for the two cores werereconstructed using weighted averaging (Birks et al.,1990a). Results Sediment core biostratigraphy and inferred pH  Dates and diatomspeciescharacteristicsof each strati-graphic zone defined in the RLGH and LDEE coresare summarized in Tables 1 and 2. Flower et al.,(1987) showed that the RLGH is strongly acidifiedwith  Brachysira vitrea  being largely replaced by aci-dobiontic  Tabellaria quadriseptata  within the last 100years. LDEE, on the other hand, is only mildly acidi-fied with frequencies of   B. vitrea  increasing and thoseof   Achnanthes minutissima  decreasing since ca. 1900.Since the 19th century, the RLGH diatom flora haschangedconsiderablymorethan that at LDEE andthisis reflected by the differences in reconstructed pH his-tories. Both lakes have been acidified by about 1 pHunit but the RLGH is and always has been more acid.The acidity change in terms of H +   concentration ismarkedly different between the two sites; the RLGHand LDEE experiencing an acidity increase of about12 and 3    g H +   l    1 , respectively (cf. Flower et al.,1987).  30 Figure 1 . The distribution of 194 potential modern analogue lakes in northwest Europe with known surface sediment diatom assemblages andwater pH. Most of the sites were part of the Surface Water Acidification Project (SWAP, see Battarbee & Renberg, 1990). Iso-lines indicatenon-marine sulphur deposition (in g m    2 yr    1 ) over the region (from Battarbee & Renberg, 1990).  Matching modern and fossil diatom assemblages Mean squared Chi-squared dissimilarities (SCD)between each diatom zone and all good and very goodanalogues are given in Tables 3 and 4. For the RLGHcore (Table 3) the pre-acidification assemblages com-prise zones four and five. Both these zones match verycloselywithsurfacesedimentdiatomsinLochTeanga,on South Uist in the Outer Hebrides, where the SCD is0.44 and 0.55 for zone 5. In zone 4, the Scottish LochsMacaterick, Tinker, Harrow and Teanga and the IrishLoughs Barra and Maumwee also have very close-ly matched assemblages (i.e. very good analogues).In the entire SWAP data set, only Loch Teanga pos-sesses a modern diatom assemblage that matches veryclosely with the RLGH zone 5 diatom assemblage.Although Loughs Veagh and Croanger in Ireland aregoodanalogues(SCD is    0.65).The veryclose matchbetweenLochTeangaandzones4and5 resultsmainlyfrom the similar proportionsof   Fragilaria virescens  v. exigua  and  Brachysira vitrea  in the modern and fos-sil samples. Frequency of another important diatom, Cyclotella kutzingiana , is relatively high in zone 5 butlow in both zone 4 and in Loch Teanga surface sedi-ment. Hence, the match is less goodbetween these lat-tertwosamples.Themorerecentcorezones1–3reflectassemblage compositions resulting from progressivelakeacidificationasacidophilousandacidobiontictaxaincreaseinabundance.Zone3assemblagesmatchmostclosely with those in modernLochs Laidonand Maca-terick where  Tabellaria flocculosa  and  Eunotia incisa are the two most abundant taxa. Zone 2 assemblages  31 Table 1 . Diatom stratigraphy of a sediment core from the Round Loch of Glenhead(southwest Scotand) summarized into five zones by cluster analysis (see text) andshowingthedepth, timeperiod, dominantdiatomtaxaandtheinferredlakeconditionsfor each zone.Round Loch of Glenhead coreZone Period Dominant taxa Lake conditlon1 1981–  Tabellaria quadriseptata  Inferred pH 4.8–4.9(0–5 cm) 1945  F. rhomboides v saxonica Eunotia incisa  Strongly acidified by Tabelleria flocculosa  atmospheric pollution  Eunotia incisa  Inferred pH 5.0–5.22 1945–  F. rhomboides v saxonica (5–11 cm) 1885  Tabellaria flocculosa  Moderately acidified by  Brachysira vitrea  atmospheric pollution3 1885  Brachysira vitrea  Inferred pH 5.2–5.3(11–16 cm) c. 1840  Eunotia incisaF. rhomboides v saxonica  Mildly acidified by Tabellaria flocculosa  atmospheric pollution  Brachysira vitrea  Inferred pH 5.3–5.64 Pre-  F. virescens v exigua (16–40 cm) 1840  F. rhomboides v saxonica  Catchment disturbance?  Eunotia incisa  loss of   Cyclotella  spp. F. virescens v exigua  Inferred pH 5.6–5.85 Pre-  Brachysira vitrea (40–80 cm) 1840  Cyclotella kuekingiana  Pristine softwater F. rhomboides v saxonica  lake  Achnanthes minutissima most closely match with Lochan Dubh but also withLochs Laidon and Macaterick.  Frustulia rhomboides var.  saxonica and  Eunotiaincisa  are the most commontaxa in these samples. Zone 1 assemblages are charac-terized by very acid tolerant taxa including  Tabellariaquadriseptata  and  T. binalis  and correspondingly thezone is matched with other strongly acidified lakes(Lochs Fleet and Valley) in the Galloway region.In the LDEE core, pre-acidificationassemblagesinzone 5 have lowest SCD values with modern diatomassemblages in Irish Loughs Claggan and Aganive.Lough Brockagh (W. Ireland), Loch White and LochHowie (Scotland) and Llyn Bodgynedd (Wales) arealso all very good analogues for LDEE zone 5 assem-blages. The two modern most closely matched lakeassemblages are currently dominated by the plankton-ic  Cyclotella kutzingiana  and periphytic  Achnanthesminutissima ,  F. virescens  var.  exigua, T. flocculosa and  B. vitrea . More recent zones 1–4 in the core representa gradual shift in assemblage composition to speciesindicative of more acid water. Zone 4 assemblagesmatch most closely with Rødlivatn(Norway)and LlynLennych (Wales). Zone 3 matches most closely withLoch Corrie nan Arr in northwest Scotland where theattached diatom  Tabellaria flocculosa  is most abun-dant and planktonic  Cyclotella  taxa are rare. Nine oth-er lakes in Britain, Ireland and Norway form goodanalogues for this zone. Zones 1 and 2 represent themost acid and recent period and  Brachysira vitrea  andother acidophilous taxa,  Eunotia incisa  and  Frustuliarhomboides  var.  saxonica , are common. These zonesmatch very closely with Lochs Harrow, Laidon, Chonand Macaterick (Scotland) and with Stor Grønningen(Norway) surface sediment assemblages.
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x