JOURNAL MENUFIND ISSUES
FIND ARTICLES
GET ACCESS
FOR CONTRIBUTORS
ABOUT THIS JOURNAL
SPECIAL FEATURES
MANEL LEIRA1, PHIL JORDAN2, DAVID TAYLOR1, CATHERINE DALTON3, HELEN BENNION4, NEIL ROSE4 andKENNETH IRVINE1DOI:&10.1111/j.06.01174.x
Journal of Applied Ecology pages 816&827, Author Information1
School of Natural Sciences, Trinity College, University of Dublin, Dublin, E-15071, I 2
School of Environmental Sciences, University of Ulster, Coleraine, BT52 ISA, UK; 3
Department of Geography, University of Limerick, Limerick, I and 4
Environmental Change Research Centre, University College London, London, WC1H 0AP, UK*Present address and correspondence: Manel Leira, Faculty of Sciences, University of A Coru&a, 15071 A Coru&a, Spain (e-mail ).Publication HistoryIssue published online: 23 JUN 2006Article first published online: 23 JUN 2006Received 6 July 2005; final copy received 6 March 2006Editor: Paul Giller
ARTICLE TOOLS
Water Framework Directive 1The European Union (EU) Water Framework Directive (WFD) requires that member states establish type-specific reference conditions for all waterbodies, including freshwater lakes. This presents a problem in those locations where human activity has resulted in significant changes to the biological, chemical and physical characteristics of waterbodies.2Seventy-six oligotrophic and meso-oligotrophic [0&19&&g total phosphorus (TP) L&1] lakes thought to be relatively unimpacted by human activity have been nominated as candidate reference lakes (CRL) by the Irish Environmental Protection Agency. This research has used palaeolimnological (lake sediment-based) techniques to test the actual and historical ecological site-specific status of a representative selection of these CRL. Where the temporal record of sedimentation was sufficiently long, the study adopted c. 1850 ad as the primary baseline date for reference conditions.3Short sediment cores were obtained from the deepest parts of 35 CRL, and chronologies were established from profiles of spheroidal carbonaceous particles (SCP). Twenty-two cores of sediment appeared to date back to c. 1850 on the basis of SCP concentrations. Unless the SCP chronologies suggested otherwise, diatom assemblages present in top and bottom samples from the cores were used as proxies of, respectively, present-day and reference conditions. Past lake water pH (DI-pH) and TP concentration (DI-TP) were inferred from the diatom data. Higher resolution sampling (up to five sediment samples per core) was adopted at seven sites where the SCP-based chronology was more robust and for at least one core from each of the most common types of CRL. Sediment chemistry data were determined to identify possible anthropogenic drivers of the observed changes in the diatom assemblages.4Ordination and dissimilarity measures identified the main patterns of variation in the diatom data. Eleven of 34 CRL for which diatom data were available showed little or no change in biological status between core top and bottom samples. Core bottom samples in six of these dated back to pre- or c. 1850 and reference (and high ecological) status could therefore be confirmed in these cases. The estimated age of core bottom samples in the remaining five cores was in the period just after c. 1850 to c. 1950 ad (four cores) or was impossible to determine (one core). Twenty-three (68%) of the CRL sampled showed biologically important deviation from the reference condition, with acidification and nutrient enrichment seemingly the main causes of change. Catchment disturbance, notably peat erosion possibly linked to recent afforestation, also appeared to have been a factor in some cases.5Synthesis and applications. This study provides the first systematic examination of changes to water quality in (perceived) pristine lakes over the last c. 150&years for Ireland, and demonstrates the potential of palaeolimnology to support the implementation of the WFD. The results indicate that diatom communities in low alkalinity lakes have been particularly altered, and acidification and nutrient enrichment appear to have been important drivers for some lakes. Furthermore, higher resolution results call into question the validity of applying c. 1850 as the date for reference conditions across Ireland.The European Union (EU) Water Framework Directive (Directive 2000/60/EC of the European Parliament and of the Council: Establishing a framework for Community action in the field of water policy) aims to maintain or improve the ecological status of the range of water resources in Europe by 2015. Accordingly, EU states are required to identify, delimit and differentiate surface waterbodies and to establish hydromorphological, physiochemical and biological type-specific reference conditions (: L327/27). Biological reference conditions, equating to high ecological status, can be established in several ways (). In Ireland, where parts of the country have relatively low levels of human activity, it is possible that some surface waterbodies show no more than minor deviations from a reference state, therefore allowing a spatial approach to the determination of biological reference conditions, at least in part.The Environmental Protection Agency (EPA) in Ireland nominated 76 lakes as candidate reference lakes (CRL) or lakes that possibly represent type-specific reference conditions, based largely on current levels of human activity in their catchments. The research presented here tested the status of a representative selection of these lakes using palaeolimnological techniques, and examined the influence of a range of potential drivers of ecological change. Because of the large number of lakes involved in the current study, a &top and bottom& approach was applied to the analysis of sediment cores. This approach involves the analysis of two samples per sediment core per site () and assumes that the top and bottom samples in a sediment core integrate, respectively, conditions at the time of coring and site-specific reference conditions, although the latter is dependent on the rate of sediment accumulation. The remains of diatoms were used as the main proxy of aquatic biological conditions, because of their sensitivity to a wide variety of aquatic variables, including acidity and nutrient availability, and their often abundant preservation in lake sediments (; ).It has been generally agreed in the UK that c. 1850 ad is a suitable reference date for the assessment of anthropogenically driven aquatic impacts (; ; ) and this date was adopted here, where it could be established. In many parts of north-western Europe, however, profound human impacts pre-date the mid-19th century (e.g. in D ), and this was certainly the case in parts of Ireland where rural population densities were often far greater than current levels during the early to mid-19th century (; ). High levels of fertilizer applications were not a feature of agriculture in Ireland until intensification commenced in the 1950s (), however, and c. 1950 ad is probably an appropriate secondary reference baseline in those locations where nutrient enrichment is potentially a significant problem.fieldwork and coringSediment cores were collected from 35 CRL from the most populated CRL typology classes. The CRL typology, which was in effect a working typology specific to Ireland, was based upon measurements of alkalinity (&&20&mg/L CaCO3, 20&100&mg/L CaCO3,&&&100&mg L&1 CaCO3), average water depth (&&&&4&m) and lake area (&&&&50&ha). All were oligotrophic or meso-oligotrophic and most were located in the western part of Ireland at altitudes of less than 300 m a.s.l. ( and ). The 35 CRL cored were divided amongst eight of the total of 12 typology classes identified by the Irish EPA: (1) low alkalinity, shallow, (2) low alkalinity, shallow, (3) low alkalinity, deep, (4) low alkalinity, deep, (6) moderate alkalinity, shallow, (8) moderate alkalinity, deep, (10) high alkalinity, shallow, and (12) high alkalinity, deep, large lakes.Figure&1. Location of the 35 CRL.Table&1.&
Summary of locational information and chemical-physical characteristics for 35 CRL cored in the present study. Water quality data were provided by the EPA and are annual means Lake nameIrishgrid ref.Lake codeTypology classAltitude(m a.s.l.)Lake area (ha)Max. depth (m)pHConductivity(&S cm&1)Alkalinity(mg L&1 CaCO3)TP&g L&1% land cover in catchmentURBANFORESTRYPASTUREAGRICULTBOGSOTHERAnnaghmoreM 900&837ANN10&46&53&1&5&78&46351159&4&60&00&0&0091&44&0&00&&0&00&8&56ArderryL 995&457ARD&4&37&81&111&66&33&846&14&60&00&9&51&0&00&0&00&90&49&0&00Ballynakill (Gorumna)L 856&225BAL&6&13&23&916&47&124420&02&5&&&&&&BaneN 550&712BAN12112&75&416&98&43297132&5&50&00&0&0094&94&4&68&&0&00&0&39BarfinnihyV 850&768BAF&3249&13&616&76&84&564&2&40&00&0&00&0&00&0&00&95&30&4&70BarraB 935&120BAR&2&90&62&6&6&06&31&543&80&50&00&0&00&1&78&0&00&62&0436&18BunnyR 375&967BUN10&17102&911&68&47361156&2&50&00&0&0036&70&0&74&&1&2261&33CloonaghlinV 610&709CLO&4109127&729&46&82&622&0&50&00&0&00&0&00&0&00&70&4129&59CullaunR 315&905CUL12&16&49&720&18&40393172&0&60&00&0&0072&15&5&44&&0&4122&01DanO 150 40DAN&4200102&933&55&11&42&0&1&60&00&8&75&1&01&0&00&51&7938&45DooC 359&394DOO&3283&&9&0&6&85&88&78&12&05120&00&0&00&0&00&0&00100&00&0&00DunglowB 782&117DUN&2&13&61&2&6&15&7310059&63&60&00&0&00&1&31&4&37&93&83&0&49EaskyG 442&225EAS&2180119&211&06&53&484&04&70&00&0&00&0&00&0&00100&00&0&00Fad Inishowen EastC 539&439FAD&3233&12&313&66&35&80&95&02&70&00&0&00&0&00&0&00&49&0850&92FeeL 790&613FEE&4&47173&731&56&55&623&06&90&0014&02&0&00&0&00&72&4613&51FeeaghF 965&000FEA&4&11394&843&07&39&869&60&80&0022&69&0&10&1&62&63&9511&65Keel (Rosses)B 847&162KEE&1136&11&410&55&31352&4&80&00&0&00&0&00&0&00&99&91&0&09KiltoorisG 676&972KIL&6&&7&43&513&57&1820527&43140&00&0&0047&8710&73&17&9723&42KindrumC 185&430KIN&8&&8&60&811&08&2731869&47110&00&0&0018&4922&16&59&34&0&00KylemoreL 770&552KYL&4&35132&225&16&59&726&99&60&0011&87&0&00&0&24&66&6121&28LeneN 510&685LEN12&93416&219&78&46250104&9&60&00&0&0078&9211&62&&0&00&9&46McNeanH 040&400MCN&8&50977&816&97&6011623&6170&0013&3926&3620&64&22&7216&89MuckanaghR 370&925MUC12&17&96&117&88&53462208&6&50&00&0&0055&29&9&18&21&7413&79NahasleamL 971&244NAH&1&33&28&1&1&46&5100&89&59&70&00&5&86&0&00&0&00&92&75&1&39NambrackkeaghL 821&603NAB&1&65&&6&7&8&85&981012&26100&0044&21&0&00&0&00&53&32&2&47NaminnC 396&419NAM&1150&15&0&7&86&551127&0100&00&0&00&0&00&0&00100&00&0&00NaminnaR 176&710NAN&1169&20&2&8&46&02&770&7&80&0036&55&0&00&0&00&63&45&0&00O&FlynnM 585&795OFL10&77137&5&3&48&51333138&9100&52&0&0054&12&0&00&42&04&3&32OoridL 930&460OOR&4&45&60&512&06&40&658&06&70&00&3&10&0&00&4&73&92&17&0&00ReaM 615&155REA12&81301&120&98&54308128&5&63&07&0&0086&7310&20&&0&00&0&00ShindillaL 960&460SHI&4&38&70&223&06&45&736&17&40&00&5&69&0&00&0&00&94&31&0&00TaltG 398&150TAL&8130&97&323&08&0119085&09&80&00&0&8126&12&0&00&73&07&0&00TayO 160 75TAY&4250&50&032&85&12&40&0&3&80&00&0&58&0&00&0&00&59&0940&34UpperV 900&817UPE&4&18169&936&16&41&582&8&50&00&6&30&0&44&2&50&83&72&7&04VeaghC 022&215VEA&4&40260&928&06&30&332&16&00&25&3&15&0&00&0&00&65&1931&41Sediment cores were collected during the period June&September 2003 from the deepest part of each CRL using a 0&5-m Renberg gravity corer (), with coring usually following an extensive bathymetric survey. Water samples were also obtained at each coring site for subsequent chemical and biological characterization, and any obvious signs of human impact were noted. Information on lake water quality and land cover in the catchment for each CRL sampled, in the form of 1990 CORINE Land Cover data, was also obtained from the EPA.laboratory analysesSediment cores were subsampled in the field at 0&5-cm intervals for the upper 5&cm and at 1-cm intervals thereafter, and bagged in labelled zip-lock bags. Concentrations of spheroidal carbonaceous particles (SCP) in samples of lake sediments provide a record of the impact of fossil fuel combustion in the region, and down-core variations in these have been shown to provide a reliable dating method (). The start of the SCP record in lake sediments at many sites across Europe is c. 1850, while c. 1950 marks the start of a rapid increase in concentrations, resulting from increased electricity generation and the widespread availability of cheap oil (), which peaked some two to three decades later. Separate studies in the north-west and north of Ireland dated the peak in SCP concentrations to 1981&& 2 and 1980&& 3 ad (). In the current study, following the method of ), SCP from five samples per core were counted at&&400 magnification under a light microscope and concentrations expressed as numbers of SCP per gram dry mass of sediment (g DM&1). Estimates of sediment age and accumulation rates based on such coarsely resolved analyses have to be treated with caution. However, the estimated accumulation rate for Dan (53&23&N, 6&16&E) corresponds closely with published 210Pb-based data for the same site (), while the accuracy of the SCP-based chronology for Nambrackkeagh (53&34&N, 9&46&E) was confirmed using recently available 210Pb-based accumulation rate data (E. McGee, unpublished data). Chronological control in the current study was based on the start of the SCP record, 1850&&&25 years (c. 1850); the rapid increase in SCP concentrations, 1950&&&10 (c. 1950); and the peak in SCP concentrations, 1980&& 3 (c. 1980) (). In addition, the top 0&5-cm thick slice of sediment in a core (the core top sample) was assumed to date to the year of coring (i.e. 2003).Samples were prepared and analysed for diatoms using standard methods (). At least 300 valves were counted for each sample using oil immersion objective and phase contrast microscopy. The relative abundance of all species (including unidentified forms) was determined as the percentage of the total count (). Diatoms were identified using standard floras ().Two samples (top and bottom) per core were analysed for 28 a higher resolution (four or five samples per core, including top and bottom samples) was adopted for cores where there was particular interest in the magnitude, rate and direction of change from the reference sample and where the SCP-based chronology was most robust (seven sites). At least one example from each of the main types of CRL was analysed at higher resolution. Generally, core samples contained abundant, well-preserved diatoms, except from the higher alkalinity lakes Bane, Cullaun, Lene, McNean and Rea. In those cases the stratigraphically contiguous sample was analysed in its place. Only in one case (Fad East, Donegal) were fossil diatoms absent throughout the core from below the upper 2&cm.Dry weight, density and the concentrations of a range of elements, including total phosphorus (TP), sodium (Na), potassium (K), calcium (Ca), iron (Fe) and manganese (Mn), were determined on core samples from all 35 CRL. Levels of sedimentary TP are used in the interpretation of nutrient enrichment as inferred from biological analyses (), while those of Na, K and Ca could indicate periods of catchment erosion. Fe and Mn in sediments are indicators of fine sediment ingress to lakes but can also be vectors for mobilization of TP in anoxic sediments ().In the majority of cases, wet density was determined from the final weight of extruded sediments of known volume. Dry weight was determined on each whole extruded sediment slice (105&&C for 24&h; ). Sediment chemistry in each core was determined in the majority of cases on the top five 0&5-cm and the bottom five 1-cm thick slices. Additional sediment samples were analysed between the uppermost and lowermost sets of samples in cores of sediment that had relatively robust SCP-based chronological control and that represented the most common types of CRL encountered (32 CRL in total).Following ), dried sediments were disaggregated by pestle and mortar and, after further drying, 0&3 g of the dried sediment was added to Teflon beakers. Samples were sequentially digested with concentrated hydrofluoric (HF), nitric (HNO3) and perchloric (HClO4) acids at temperatures from 40&&C to 60&&C and preserved as acidified 25-mL 30 samples were digested at a time and a blank, certified reference material (CRM) and repeat digest were included for quality control in each run of measurements. Chemical concentrations (mg g&1) were determined on 20& dilutions in diluted HNO3 using an Inductively coupled plasma & Optical emission spectroscopy (ICP-OES), and were also transformed to an accumulation rate (mg cm&2 year&1) form using SCP-based estimates of sediment accumulation rate and measurements of sediment dry density.data analysesAnalyses of diatom data involved samples from 34 CRL but, as reliable chronologies could not be established for lakes Lene, O&Flynn and Tay on the basis of SCP concentrations, discussions concerning down-core variations in sediment chemistry are therefore restricted to 31 CRL. The degree of floristic change between diatom assemblages between top and bottom sediment samples was assessed using the squared chord distance dissimilarity index (SCD) as used in recent studies on diatom assemblages (; ). SCD emphasizes the pattern in the data and has been shown theoretically to perform better than other coefficients for determining ecological resemblance amongst samples (). SCD scores range from 0 to 2, with 0 indicating identical species composition and 2 entirely different. An SCD score & 0&4 (approximating to the 2&5 percentile) was used in the current research to define sites with low floristic change between the bottom and top sample. This is more stringent than the 5th percentile used by ) and reflects revised thinking about what constitutes biologically important change at a site (Environmental Change Research Centre, University College London, London, UK). Unimpacted lakes in this database typically have SCD scores of & 0&4.Detrended correspondence analysis (DCA; ) was used to identify the main patterns of variation in the diatom data, and to establish the directions and magnitude of changes in biological conditions at each coring location. Version 4&5 of canoco was employed (). Prior to all analyses, diatom abundances were square-root transformed in order to stabilize variance, and rare species were down-weighted.Diatom-inferred pH and TP (DI-pH and DI-TP) were established using standard weighted averaging (; ; ; ) to quantify any changes in nutrient status and acidity. All diatom-inferred values were established using the computer software package C2 (). In the absence of a diatom assemblage&water quality training set specific to Ireland, DI-pH and DI-TP were determined using a preliminary training set that comprised diatom counts for surface sediment samples and several years of measurements of lake water quality made by the EPA. The training set for pH comprised information from the 35 CRL studied in
pH values ranged between 5&11 and 8&54 and had a median value of 6&5 pH units. The resultant weighted averaging partial least-squares two-component (WA-PLS2) model (r2&=&0&84) had a root mean squared error of prediction (RMSEP) of 0&43 pH units. The training set for TP included additional information from 10 nutrient-enriched lakes in Ireland in order to extend the TP gradient. TP values in the training set were in the range 0&2&55 log&10&&g TP L&1 and had a median value of 0&90 log&10&&g (8&0&&g) TP L&1. The resultant WA-PLS2 model (r2&=&0&64) generated a RMSEP = 0&202 log&10&&g TP L&1. The difference between current DI-TP and reference DI-TP was used to derive a qualitative estimate of degree of change, with a difference in DI-TP greater than the RMSEP deemed biologically important.A comparison of the models used in the current research to derive pH and TP from diatom assemblages with other transfer functions recently developed for Northern Ireland and Europe is presented in . In general, the statistical performance of the predictive models used in the current work compared well with other diatom TP and pH transfer functions.Table&2.&
Summary performance statistics of published diatom TP and pH transfer functions, showing the statistics of the Irish models for comparison. The RMSEP are based on either leave-one-out jack-knifing (jack) or bootstrapping (boot) Training set referenceNumber of lakesEstimationPredictionRMSEr2RMSEPr2Irish lakes pH; this paper&350&2820&930&427 (jack)0&86Irish lakes TP; this paper&450&1740&840&202 (jack)0&64Surface water acidification project (SWAP) pH; Birks et&al. (1990)1670&2300&910&290 (jack)0&86Acidification of mountain lakes: Palaeolimnology and ecology (ALPE) pH; 1180&1330&970&326 (jack)0&82North-west Europe TP; 1520&1500&910&210 (jack)NANorth Irish TP; &430&1720&75NANASouth-east England TP; &300&1600&790&280 (boot)NAapproximate sediment chronologies for crlDown-core variations in SCP concentrations (see Appendix S1 in the supplementary material) permitted allocation of each of the 35 sediment cores analysed to one of three categories, based on the apparent completeness of the SCP profile: (1) complete, where the start of the SCP record was within the length of the core and depths for c. 1950 and c. 1980 were extrapolated assuming a constant sediment accumulation rate (13 cores in total); (2) curtailed, where SCP was only present in the uppermost part of the core, presumed to be because of slow sediment accumulation rates rather than a loss of sediments (11 cores in total); and (3) incomplete, where SCP was present in all samples and an estimate for c. 1850 was extrapolated to a sediment depth below the base of the core (11 cores in total). It was thus possible to estimate the age of the bottom sediment core sample in 18 cases, with rates of sediment accumulation in these 18 cores varying widely, from a low of 0&09&cm year&1 (Dunglow) to a high of 0&58&cm year&1 (Kiltooris) ().Table&3.&
Approximate chronologies and estimated sediment accumulation rates, based on down-core variations in SCP abundances LakeTypology classLength of core (cm)Estimated depth1850&&&25 (cm)Estimated depth1950&&&10 (cm)Estimated depth1980&& 5 (cm)Estimated age of bottom diatom sample analysedEstimated sedimentaccumulation rate (cm year&1)Annaghmore102017&195&72&318340&12Arderry&436&&9?&&&&Ballynakill&63629&3510&134&5&518310&21Bane12174&5&8?1&5&3?0&5&1?&&Barfinnihy&340&&9?&&&&Barra&23650&6317&227&5&919050&37Bunny102521&247&8&53&418330&15Cloonaghlin&4349&17?3&6?1&2&5?&&Cullaun12318&14?2&5&5?1&2&5?&&Dan&43143&5715&206&5&8&519090&33Doo&33172&8625&3010&5&1319440&53Dunglow&21113&154&62&318850&09Easky&22723&278&93&418380&16Fad&327&&276&72&4&&Fee&43810&18?3&5&6&5?1&5&3?&&Feeagh&43666&7623&2710&1219260&47Keel&14128&3011&153&4&517880&19Kiltooris&63466&11023&4010&1919440&58Kindrum&83020&227&83&417760&13Kylemore&440&&10?&&&&Lene1229&&29&&&&McNean&81912&143&5&51&5&2&519500&36Muckanagh124231&40?10&5&14?4&5&6?&&Nahasleam&127&58&14?2&5&5?1&2&5?&&Nambrackkeagh&126&517&216&72&5&3&517890&12Naminn&13022&246&72&5&3&518040&15Naminna&126&58&15?2&5&5&5?1&2&5?&&O&Flynn1042&&42&&&&Oorid&439&&10?&&&&Rea12369&17?3&6?1&2&5?&&Shindilla&42918&216&72&5&3&517750&13Talt&82218&215&5&72&5&3&518300&13Tay&436&&36&&&&Upper&440&535&4011&144&5&618400&25Veagh&43460209&&down-core variations in sediment chemistry and diatom dataSediment chemistry data, in concentration form, together with density and dry weight measurements, are presented in Appendixes S2 and S3, respectively (see the supplementary material). Differences in accumulation rates between mean reference and mean present-day samples for the three cations Ca, K and Na are shown in . In general, the results did not indicate increasing exogenic inputs, with only Talt showing increases in Ca, Na and K accumulation rates, which was consistent with increased sediment dry weight towards the top of the core. Stable or falling exogenic inputs of sediment to the CRL studied were also evident when the sediment chemistry data were expressed in concentration form. Changes in Mn and Fe accumulation rates are shown in . Manganese accumulation rates increased in some lakes where there were no other cation increases, which is indicative of mobilization and accretion at a redox boundary (). Increases in Mn were also apparent when the data were expressed in concentration form. Again, Talt was an exception, where Mn and Fe increases were also likely to be dependent on exogenic inputs consistent with increased dry weights. Total P accumulation rates () indicated a decrease from reference rates for all lakes, with the exception of Talt and Cullaun.Figure&2. (a) Sodium accumulation rates. (b) Calcium accumulation rate. Calcium in the six high-alkalinity lakes is an order of magnitude higher than other lakes and reflects the predominance of calcareous soils in lake catchments. (c) Potassium accumulation rates.Figure&3. Sediment chemistry accumulation rates showing the reference and present-day (c. 2003) samples and the mean of the top five samples (0&2&5&cm) for the 32 CRL sampled for which there was tightest chronological control. &Mean REF& refers to the mean of the samples encompassed by the SCP dating limits. (a) Manganese accumulation rates. (b) Iron accumulation rate.Figure&4. Total phosphorus accumulation rates showing the reference and present-day (c. 2003) samples and the mean of the top five samples (0&2&5&cm) for the 32 CRL sampled for which there was tightest chronological control.The main down-core differences in diatom assemblages are tabulated in Appendix S4 (see the supplementary material). SCD scores between diatom assemblages in present-day and bottom core samples ranged from 0&046 to 1&769 (). In order to facilitate assessment of the departure from reference conditions, core bottom, present-day and, in some cases, mid-core samples from each lake were combined on the same DCA ordination biplot (). Eleven (32%) of the 34 sites for which top&bottom comparisons in diatom assemblages were possible generated SCD scores & 0&4, indicating little or no change in biological status when compared with the reference benchmark. These 11 cases were distributed among the different CRL typology classes, although none were in typology class 12 (deep, large, high alkalinity). Aside from CRL in typology class 12, where differential preservation of diatom frustules may have been a factor influencing the results, large, deep, low alkalinity lakes (typology class 4) were most consistent in showing biologically important changes compared with reference conditions. Eighty-two percent of CRL within this class generated SCD scores & 0&4. Many of the moderate to high alkalinity CRL (typology classes 6&12) studied had also experienced important biological changes, based on the SCD scores, although at some of these differential preservation of diatom frustules may have been an important factor.Table&4.&
Summary of down-core variations in biological and inferred parameters for the 35 CRL studied (full names and lake codes shown). Data comprise squared chord distance (SCD) scores for pairs of reference and surface samples from 34 of the 35 CRL, along with Hill's SD units and diatom-inferred TP and pH changes (DI-pH and DI-TP). &Prop. change in DI-TP&,&the proportion of change in DI-TP in the c. 2003 sample compared with reference, where, for example, 0&8&=&20% decrease, 1&2&=&20% increase, etc. CRL that appear on the basis of their SCD scores (i.e. SCD scores & 0&4) to have had their CRL status verified in this study are in bold. No SCD score based on an intracore comparison was calculated for Fad Inishowen East (Fad East) (County Donegal, Lake code FAD) because of an absence of well-preserved diatoms throughout the core (see text for further details) LakeLake codeTypology classSCDSD Hill's unitsChange in DI-pH unitsProp. change in DI-TPAnnaghmoreANN100&8351&1180&110&95ArderryARD&40&8591&797&0&081&31BallynakillBAL&60&5091&24&0&091&09BaneBAN120&5571&1120&211&01BarfinnihyBAF&30&1390&7310&031&04BarraBAR&20&4090&986&0&011&01BunnyBUN100&3510&940&030&90CloonaghlinCLO&41&5991&4160&391&31CullaunCUL121&3021&376&0&291&18DanDAN&40&4100&806&0&430&96DooDOO&30&2591&139&0&211&02DunglowDUN&20&1711&176&0&080&97EaskyEAS&21&0341&1&0&471&08FadFAD&3&&&&FeeFEE&40&9671&494&0&092&05FeeaghFEA&41&7691&822&0&191&89KeelKEE&10&3191&1&0&141&31KiltoorisKIL&60&2881&171&0&021&03KindrumKIN&80&4231&260&251&25KylemoreKYL&41&0441&551&0&760&91LeneLEN120&4781&0010&061&01McNeanMCN&80&1481&0560&151&14MuckanaghMUC120&4580&884&0&121&10NahasleamNAH&10&2960&787&0&101&03NambrackkeaghNAB&10&750&9840&121&25NaminnNAM&10&5621&205&0&051&09NaminnaNAN&10&7351&123&0&300&92O&FlynnOFL100&3271&0450&190&83OoridOOR&40&9871&715&0&191&08ReaREA120&4721&554&0&251&02ShindillaSHI&40&4161&0390&101&20TaltTAL&80&961&508&0&250&99TayTAY&41&1691&069&0&580&58UpperUPE&40&0461&019&0&090&98VeaghVEA&40&2450&9230&051&01Figure&5. DCA plot combining reference and present-day (c. 2003) sediment samples, as well as mid-core samples where available. Lines connect the reference and present-day samples for each core. The trajectory is the direction of floristic change and its length is a measure of floristic difference (units = Hill's SD). See
for explanation of lake codes.Down-core differences in DI-pH and DI-TP are given in . These results must be viewed in the context of the limitations in the models used to infer pH and TP based on diatom and environmental data for lakes in Ireland. Because of the known deficiencies in the model used, differences in DI-TP between core top and bottom samples are also reported in
as proportional change. Present-day reconstructions of pH at 12 of the CRL sampled showed higher pH and 22 showed lower pH compared with core bottom samples, although only four lakes (Dan, Easky, Kylemore and Tay, all low alkalinity CRL) showed a decline in pH & RMSEP. Ten CRL appeared to have experienced a reduction in DI-TP, while DI-TP increased at 24 sites. However, only in Tay did the reduction in DI-TP appear important (& RMSEP). In comparison, two low-alkalinity CRL showed an increase in DI-TP & RMSEP between core bottom and present-day conditions: Fee and Feeagh (both EPA typology class 4). Arderry and Cloonaghlin also exhibited a change in DI-TP, although no relevant and concurrent increases in TP accumulation rate were evident.Of the seven cores studied at higher resolution, low alkalinity lakes showed the most striking changes in DI-pH and DI-TP (e.g. Easky, Keel and F see Appendix S5 in the supplementary material). Keel and Fee exhibited increases in DI-TP up to c. 9&&g L&1 TP, while at Keel the increase in DI-TP occurred before c. 1850 and showed some stability since the reference date. Ballynakill (typology class 6, moderate alkalinity) and Annaghmore and Muckanagh (respectively, typology classes 10 and 12, high alkalinity) showed only relatively small changes in DI-pH and DI-TP between the reference and present-day samples.Insignificant differences between core top and bottom diatom assemblages were found in 11 CRL (32% of those studied), providing support for largely stable ecological conditions at these sites. Sedimentary P accumulation rates largely showed a decrease from reference. Reduced rates of P accumulation compared with reference rates may be an artefact of the data. However, they may also reflect changes in land use and human population levels following the famines in Ireland of the early to mid-19th century, which had their most profound impact (rural depopulation on a massive scale) in the west of the country (). The core bottom sample in six of the lakes represent c. 1850 or earlier reference conditions, and therefore reference (and high ecological) status could be verified in these cases. Core bottom samples for the other five CRL dated to the period between the primary and secondary reference baselines (i.e. c. 1850&c. 1950). As this period was prior to agricultural intensification and afforestation in much of Ireland, the reference status of these sites would also appear justified.Biologically important changes were evident between reference and present-day conditions for 68% of the CRL studied. The main drivers of change appear to be either nutrient enrichment or increased acidity. Phosphorus transfer to freshwater lakes is a major cause of nutrient enrichment in Ireland at present. No link was found in the current research between indicators of the mass transfer of exogenic soil material and those lakes with increasing TP accumulation, suggesting that increased discrete point sources, such as those associated with rural dwellings and septic tanks, and diffuse soluble P losses from P-saturated soils or inappropriate slurry spreading (during wet weather) (), are possible drivers of changes in the diatom assemblages recorded here.It is likely that increased TP accumulation rates at Talt are linked to evidence of inwash, while a link between increased Ca and TP accumulation rates at Cullaun may be the result of co-precipitation of TP with CaCO3 (). There is, however, a caveat on the interpretation of these sediment chemistry data. First, the calculation of dry mass sediment accumulation rates using SCP chronologies may introduce an error because the model assumes a linear accumulation rate between dated horizons. The level of error, however, is likely to be minimized in the current research because of the relatively low levels of human activity (and therefore rates of erosion) in the catchments for many of the CRL cored. Secondly, and especially with regard to the sedimentary P profiles, chemicals may become concentrated at the sediment&water interface as a result of diagenesis and mobility. According to the concentration data, this does not appear to be a problem, however. Even in the low alkalinity lakes, where diagenetic mobilization would be expected to occur (as a result of exogenic Fe and Mn inputs from acid soils) and where changes in the Fe and Mn profiles indicate mobilization, only a minor concurrent increase in the sedimentary P concentration profiles is evident. Except for Arderry and Fee, by far the largest increases in sedimentary P concentration are in the high alkalinity lakes, where there are no indications of Fe and Mn mobilization and where these would be supply limited from calcareous catchment soils.Peat inwash could have influenced the composition at several lakes that have experienced biologically important changes in diatom assemblages, but show no major change in the DI-pH and DI-TP profiles. CRL appearing to have been impacted by peat inwash experienced a shift to a benthic diatom assemblage characterized by Fragilaria taxa and, in some cases, the loss of planktonic assemblages. Small, benthic Fragilaria are considered to be pioneering assemblages and have been related to rapidly changing environmental conditions (e.g. increased turbidity in the water column). Furthermore, diatom community changes associated with peat inwash causing a tangible impact on lake sediments are usually characterized by the loss of planktonic assemblages (). Peat erosion could have been linked to afforestation, as conifer plantations, many dating to the period post-1950, accounts for a substantial proportion of the land cover in the catchments for CRL that may have experienced peat erosion.Of particular note is the comparatively high incidence of important alterations in biological conditions among the low alkalinity, large deep lakes (82% of CRL in typology class 4 generated SCD scores & 0&4). Nutrient enrichment does not appear to have been the driver here as increases in DI-TP are relatively minor and are not generally supported in the sedimentary TP data. It could be, however, that biologically important TP concentrations in lake water are not in balance with sediment-based TP. Moreover, relatively small increases in TP concentrations in P-limited lakes may impact diatom populations before any increases are evident in sediment chemistry: rapid DI-TP change has been found in some historically oligotrophic lakes in Northern Ireland prior to increases in sediment TP (P. Jordan & N. J. Anderson, unpublished data). Acidification appears to have been a factor at several of these lakes, however, notably Dan, Kylemore and Tay.The more finely resolved sediment chemistry and diatom data from Keel (low alkalinity) and Talt (moderate alkalinity) are of particular interest. The sedimentary data for Keel indicate relatively little change between c. 1850 and the present-day. The rate of sedimentation at Keel was, however, low (0&19&cm year&1) and the base of the 41-cm long core of sediment obtained dated to the late 18th century, based on an extrapolation of the SCP chronology. Comparisons of diatom assemblages and DI-TP between the core bottom and present-day samples from this site indicate much more important changes than those between c. 1850 and present-day, because of increased abundances of Asterionella ralfsii (a taxon that was not recorded in pre-c. 1850 sediments). The ecology of this taxon is not well known but appears to be indicative of peatland disturbance and nutrient enrichment (). It could therefore be argued that c. 1850 is not an appropriate reference date for Keel, and this may also be the case for several other lakes in catchments that until the mid-19th century supported far higher human population densities than today. The results from Talt indicate a pattern of change that may be hidden in some of the less finely resolved core data sets. The sediment chemistry for this site indicates catchment disturbance in the form of increases in sediment dry weight and in Ca, Fe, K, Mn, Na and TP accumulation rates between c. 1950 and the present-day samples. According to the dry weight and sediment chemistry data, the catchment has stabilized since this disturbance event, although diatom communities and pH levels have yet to recover fully.Analysis of the moderate and high alkalinity lakes (typology classes 6, 8, 10 and 12), where dissolution of silica diatom frustules has caused preservation and therefore interpretation problems, highlights the difficulties of using diatoms to assess floristic change in these systems. As the sedimentary data that are available for sites within these classes show that some deviation from reference conditions may have occurred, a separate study should focus specifically on these typology classes using biological indicators other than diatoms (e.g. chironomids, cladocera, ostracods and pigments). A further limitation of this investigation was that it was only possible to consider in any detail a restricted range of potential causes of variability, despite the potential role of other important influencing factors, notably climate change (; ; ; ; ; ; ). A number of lakes in this study have experienced a rise in various Cyclotella taxa compared with reference conditions, which have been linked to climate change throughout the Canadian arctic () as well as in temperate regions ().To conclude, results generated from this first, systematic palaeolimnology-based examination of the recent ecological histories of oligotrophic and meso-oligotrophic lakes in Ireland highlight the potential of palaeolimnological studies to assess reference conditions in aquatic ecosystems. Thirty-two per cent of CRL studied show relatively little deviation from reference conditions, thus confirming the reference status of these 11 lakes. Large, deep, low alkalinity lakes in particular (82% of this type of CRL studied) show biologically important deviation from reference conditions, and acidification and nutrient enrichment are suggested as important drivers of recent changes, although climate change may also be important. Finally, the data suggest that c. 1850 may not be an appropriate reference baseline in all cases, especially when viewed in the context of the history of rural Ireland since the mid-19th century.The research presented here was supported by the EPA (project number 2002-W-LS/7). Thanks are due to Jim Bowman of the EPA, Eddie McGee for the 210Pb data from Lough Nambrackkeagh, Sheila McMorrow for assistance with the figures, and to several staff and postgraduate research students from UU Coleraine, Trinity College, University of Dublin and University of Limerick, most notably Richard McFaul and Guangjie Chen, for assistance with fieldwork. Finally, thanks are owed to the numerous landowners who facilitated access to CRL and to the referees of an earlier version of this paper for their very helpful and constructive comments.
Andersen, J.H., Conley, D.J. & Hedal, S. (2004) Palaeoecology, reference conditions and classification of ecological status: the EU Water Framework Directive in practice. Marine Pollution Bulletin, 49, 283&290. ,,
Anderson, N.J., Rippey, B. & Gibson, C.E. (1993) A comparison of sedimentary and diatom-inferred phosphorus profiles: implications for defining pre-disturbance nutrient conditions. Hydrobiologia, 253, 357&366. ,
Battarbee, R.W. (1999) The importance of palaeolimnology to lake restoration. Hydrobiologia, 395/396, 149&159. ,
Battarbee, R.W., Jones, V.J., Flower, R.J., Cameron, N.G., Bennion, H., Carvalho, L. & Juggins, S. (2001) Diatoms. Tracking Environmental Change Using Lake Sediments. Volume 3. Terrestrial, Algal, and Siliceous Indicators (eds J.P.Smol, H.J. B.Birks & W.M.Last), pp. 155&202. Kluwer Academic Publishers, Dordrecht, the Netherlands.
Bennion, H. (1994) A diatom-phosphorous transfer function for shallow, eutrophic ponds in southeast England. Hydrobiologia, 275/276, 391&410. ,
Bennion, H., Fluin, J. & Simpson, G.L. (2004) Assessing eutrophication and reference conditions for Scottish freshwater lochs using subfossil diatoms. Journal of Applied Ecology, 41, 124&138. Direct Link:
Bennion, H., Juggins, S. & Anderson, N.J. (1996) Predicting epilimnetic phosphorous concentrations using an improved diatom-based transfer function and its application to lake eutrophication management. Environmental Science and Technology, 30, 2004&2007. ,
Birks, H.J.B., Juggins, S. & Line, J.M. (1990) Lake surface&water chemistry reconstructions from palaeolimnological data. The Surface Waters Acidification Programme (ed. B.J.Mason), pp. 301&313. Cambridge University Press, Cambridge, UK.
Birks, H.J.B., Line, J.M., Juggins, S., Stevenson, A.C. & Ter Braak, C.J.F. (1990) Diatoms and pH reconstruction. Philosophical Transactions of the Royal Society of London Series B, Biological Sciences, 327, 263&278. ,,
Boyle, J.F. (2001) Inorganic geochemical methods in palaeolimnology. Tracking Environmental Change Using Lake Sediments. Volume 2. Physical and Geochemical Methods (eds W.M.Last & J.P.Smol), pp. 83&141. Kluwer Academic Publishers, Amsterdam, the Netherlands.
Ter Braak, C.J.F. & Šmilauer, P. (2002) Canoco 4&5. Biometris, Wagenigen, the Netherlands.
Bradshaw, E.G. (2001) Linking land and lake. The response of lake nutrient regimes and diatoms to long-term land-use change in Denmark. PhD Thesis. University of Copenhagen, Copenhagen, the Netherlands.
Cameron, N.G., Birks, H.J.B., Jones, V.J., Berge, F., Catalan, J., Flower, R., Garcia, J., Kawecka, B., Koinig, K.A., Marchetto, A., S&nchez-Castillo, P., Schmidt, R., Šiško, M., Solovieva, N., Štefkov&, E. & Toro, M. (1999) Surface-sediment and epilithic diatom pH calibration sets for remote European mountain lakes (AL: PE project) and their comparison to the Surface Waters Acidification Programme (SWAP) calibration set. Journal of Paleolimnology, 22, 291&317. ,
Davison, W. (1993) Iron and manganese in lakes. Earth-Science Reviews, 34, 119&163. ,,
Donnelly, J.S. (2001) The Great Irish Potato Famine. Boston Publishing, Stroud, UK.
Hill, M.O. & Gauch, H.G. (1980) Detrended correspondence analysis, an improved ordination technique. Vegetatio, 42, 47&58. ,
Hilton, J., Lishman, J.P. & Millington, A. (1986) A comparison of some rapid techniques for the measurement of density in soft sediments. Sedimentology, 33, 777&781. Direct Link:
Huang, C.C. & O&Connel, M. (2000) Recent land-use and soil-erosion history within a small catchment in Connemara, western Ireland: evidence from lake sediments and documentary sources. Catena, 41, 293&335. ,
Hudson, J.J., Dillon, P.J. & Somers, K.M. (2003) Long-term patterns in dissolved organic carbon in boreal lakes: the role of incident radiation, precipitation, air temperature, southern oscillation and acid deposition. Hydrology and Earth System Sciences, 7, 390&398. ,,
Jager, P. & Rohrs, J. (1990) Coprecipitation of phosphorus with calcite in the eutrophic Wallersee (alpine foreland of Salzburg, Austria). Internationale Revue der Gesamten Hydrobiologie, 75, 153&173. Direct Link:
Jones, D.S., Stevenson, A.C. & Battarbee, R.W. (1989) Acidification of lakes in Galloway, south west Scotland: a diatom and pollen study of the post-glacial history of the Round Louch of Glenhead. Journal of Ecology, 77, 1&23. ,
Jordan, P., Menary, W., Daly, K., Kiely, G., Morgan, G., Byrne, P. & Moles, R. (2005) Patterns and processes of phosphorus transfer from Irish grassland soils to rivers: integration of laboratory and catchment studies. Journal of Hydrology, 304, 20&34. ,,
Juggins, S. (2003) C2 Software for Ecological and Palaeoecological Data Analysis and Visualisation. User Guide, Version 1&3.
Newcastle University, Newcastle upon Tyne, UK.
Karst-Riddoch, T.L., Pisaric, M.F.J. & Smol, J.P. (2005) Diatom responses to 20th century climate-related environmental changes in high-elevation mountain lakes of the northern Canadian Cordillera. Journal of Paleolimnology, 33, 265&282. ,
Korsman, T. & Birks, H.J.B. (1996) Diatom-based water chemistry reconstructions from northern Sweden: a comparison of reconstruction techniques. Journal of Paleolimnology, 15, 65&77. ,
K&ster, D., Racca, J.M.J. & Pienitz, R. (2004) Diatom-based inference models and reconstructions revisited: methods and transformations. Journal of Paleolimnology, 32, 233&246. ,
Krammer, K. & Lange-Bertalot, H. (198691) Bacillariophyceae. 1&4 Teil. S&&wasserflora Von Mitteleuropa (eds H.Ettl, J.Gerloff, H.Heynig & D.Mollenhauer). Gustav Fischer-Verlag, Stuttgart, Germany.
Liehu, A., Sandman, O. & Simola, H. (1986) Effects of peatbog ditching in lakes: problems in palaeolimnological interpretation. Hydrobiologia, 143, 417&424. ,
Lotter, A.F., Appleby, B.G., Bindler, R., Dearing, J.A., Grytnes, J.A., Hofmann, W., Kamenik, C., Lami, A., Livingstone, D.M., Ohlendorf, C., Rose, N. & Sturm, M. (2002) The sediment record of the past 200 years in a Swiss high-alpine lake: Hagelseewli (2339 m a.s.l.). Journal of Palaeolimnology, 28, 111&127. ,
Mackereth, F.J. (1966) Some chemical observations of post-glacial lake sediments. Philosophical Transactions of the Royal Society of London Series B, Biological Sciences, 250, 165&213. ,,
Michelutti, N., Douglas, M.S.V. & Smol, J.P. (2003) Diatom response to recent climate change in a high arctic lake (Char Lake, Cornwallis Island, Nunavut). Global and Planetary Change, 38, 257&271. ,,
Moss, B., Johnes, P. & Phillips, G. (1996) The monitoring of ecological quality and the classification of standing waters in temperate regions: a review and proposal based on a worked scheme for British waters. Biological Reviews, 71, 301&339. Direct Link:
Nowlan, N.V., Mitchell, P.I., Murray, D.A., Leon Vintro, L. & Seymore, E. (2000) Measurement and interpretation of lead-210 and caesium-137 profiles in relation to recent sedimentation in some Irish lakes. Verhandlungen Internationale Vereinigung Fiir Theoretische und Angewandte Limnologie, 27, 2303&2306.
OJEC (2000) Directive 2000/60/EC of the European Parliament and of the council of 23 October 2000 establishing a framework for community action in the field of water policy. Official Journal of the European Communities, L327, 1&72.
Overpeck, J.T., Webb, T. & Prentice, I.C. (1985) Quantitative interpretation of fossil pollen spectra: dissimilarity coefficients and the method of modern analogs. Quaternary Research, 23, 87&108. ,,
Renberg, I. (1991) The HON-Kajak sediment corer. Journal of Paleolimnology, 6, 167&170.
Rippey, B. & Anderson, N.J. (1996) Reconstruction of lake phosphorus loading and dynamics using the sedimentary record. Environmental Science and Technology, 30, 1786&1788. ,
Rose, N.L. (1994) A note on further refinements to a procedure for the extraction of carbonaceous fly-ash particles from sediments. Journal of Paleolimnology, 11, 201&204. ,
Rose, N.L. (2001) Fly-ash particles. Tracking Environmental Change Using Lake Sediments. Volume 2. Physical and Geochemical Methods (eds W.M.Last & J.P.Smol), pp. 319&349. Kluwer Academic Publishers, Amsterdam, the Netherlands.
Rose, N.L. & Appleby, P.G. (2005) Regional applications of lake sediment dating by spheroidal carbonaceous particle analysis. I. United Kingdom. Journal of Paleolimnology, 34, 349&361. ,
Rose, N.L., Harlock, S., Appleby, P.G. & Battarbee, R.W. (1995) Dating of recent lake sediments in the United Kingdom and Ireland using spheroidal carbonaceous particle (SCP) concentration. Holocene, 5, 328&335. ,
R&hland, K., Priesnitz, A. & Smol, J.P. (2003) Palaeolimnological evidence from diatoms for recent environmental changes in 50 lakes across the Canadian arctic treeline. Arctic, Antarctic and Alpine Research, 35, 110&123. ,
Simpson, G.L., Shilland, E.M., Winterbottom, J.M. & Keay, J. (2005) Defining reference conditions for acidified waters using a modern analogue approach. Environmental Pollution, 137, 119&133. ,,
Smol, J.P. (2002) Pollution of Lakes and Rivers: A Paleoenvironmental Perspective. Arnold Publishers, London, UK. Co-published by Oxford University Press, New York, NY.
Smol, J.P. & Cumming, B.F. (2000) Tracking long-term changes in climate using algal indicators in lake sediments. Journal of Phycology, 36, 986&1011. Direct Link:
Smol, J.P., Wolfe, A.P., Birks, H.J.B., Douglas, M.S.V., Jones, V.J., Korhola, A., Pienitz, R., R&hland, K., Sorvari, S., Antoniades, D., Brooks, S.J., Fallu, M.-A., Hughes, M., Keatley, B., Laing, T., Michelutti, N., Nazarova, L., Nyman, M., Paterson, A.M., Perren, B., Quinlan, R., Rautio, M., Saulnier-Talbot, E., Siitonen, S., Solovieva, N. & Weckstr&m, J. (2005) Climate-driven regime shifts in arctic lake ecosystems. Proceedings of the National Academy of Sciences, 102, 4397&4402. ,,,
Sorvari, S., Korhola, A. & Thompson, R. (2002) Lake diatom response to recent Arctic warming in Finnish Lapland. Global Change Biology, 8, 153&163. Direct Link:
Stoermer, E.F. & Smol, J.P. (1999) The Diatoms. Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge, UK.
Tunney, H. (1990) A note on a balance sheet approach to estimating the phosphorus fertiliser needs in agriculture. Irish Journal of Agricultural Research, 29, 149&154.
Wolin, J.A. & Stoermer, E.F. (2005) Response of a Lake Michigan coastal lake to anthropogenic catchment disturbance. Journal of Paleolimnology, 33, 73&94. ,Appendix S1: INSIGHT SCP data.
Appendix S2: Down-core variations in sediment chemistry data
Figure 1: Calcium concentrations for 32 candidate reference lake sediments. High alkalinity lakes are noted for very high sedimentary calcium concentrations.
Figure 2: Iron concentrations for 32 candidate reference lake sediments.
Figure 3: Manganese concentrations for 32 candidate reference lake sediments.
Figure 4: Potassium concentrations for 32 candidate reference lake sediments.
Figure 5: Sodium concentrations for 32 candidate reference lake sediments.
Figure 6: Phosphorus concentrations for 32 candidate reference lake sediments.
Figure 7: Certified reference materials for TP included in batch digests. Digests were repeated if CRM results were ? the confidence limits of the certified concentrations (i.e., 1.00mg/g ? 0.05mg/g).
Appendix S4
Biological changes in the study cores
Appendix S5
Diatom-inferred reconstructions and changes of sediment TP, Fe and Ca for seven cores that were analysed at relatively high resolution.
Figure 1: DI-TP reconstructions and sediment TP percentage for seven cores that were analysed at relatively high resolution.
Figure 2: DI-pH reconstructions and percentage changes in Fe and Ca accumulation rates for seven cores that were analysed at relatively high resolution.}