- Email: [email protected]
Sediment nutrient characteristics and aquatic macrophytes in lowland English rivers
The Science of the Total Environment 266 Ž2001. 103᎐112
Sediment nutrient characteristics and aquatic macrophytes in lowland English rivers Stewart J. Clarke, Geraldene WhartonU Department of Geography, Queen Mary and Westfield College, Uni¨ ersity of London, London E1 4NS, UK Received 17 September 1999; accepted 28 June 2000
Abstract Aquatic macrophytes play an important role in the nutrient dynamics of streams. As a result, there is much interest in their use as trophic indicators. However, the relationship between aquatic macrophytes and the trophic status of rivers is a complex one, partly because of the effects of a wide range of environmental variables and partly because submerged, rooted macrophytes can absorb nutrients from the river sediments andror the water column. Experiments which have tried to establish the relative importance of sediments or water as sources of nutrients are inconclusive and further work is needed to establish how sediment nutrient characteristics vary within and among rivers Žspatially and temporally. and the inter-relationships between sediment nutrients, water column chemistry and macrophytes. This paper presents the initial findings from a study of 17 lowland rivers in southern England which is exploring the spatial variability of sediment characteristics Žtotal and inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter content. and the relationship with aquatic macrophytes. The preliminary analysis indicates that although sediment characteristics are highly variable within 100-m river reaches, the variability across the 17 rivers is even greater; this is despite the limited geographic and trophic range of the study sites. The results presented in this paper also give some indication of the sediment characteristics associated with five macrophyte species but it is too early to ascribe sediment preferences for particular species. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Aquatic macrophytes; Sediments; Nutrients; Trophic indicator
U
Corresponding author. Tel.: q44-20-7882-5436; fax: q44-20-8981-6276. E-mail address: [email protected] ŽG. Wharton.. 0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 7 5 4 - 3
104
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
1. Introduction Shallow, low-energy, temperate streams, of the type found in lowland England, often have low flow velocities and fine, nutrient-rich sediments that result in abundant macrophyte growth ŽDemars and Harper, 1998; cf. Danish streams as described by Sand-Jensen, 1997.. Under these conditions, macrophytes Ždescribed as ‘biological engineers’ by Sand-Jensen, 1997. play a key role in nutrient dynamics by: trapping organic matter and debris ŽSand-Jensen, 1998.; acting as temporary sinks for nutrients during periods of growth Žcf. Canfield et al., 1983.; and functioning as a link between the sediment and water column through the release of nutrients, for example from senescent and decaying stems and leaves ŽCarignan and Kalff, 1980. or from the sediment through the creation of anoxic sediments within dense plant stands. As a result, there is much interest in aquatic macrophytes as trophic indicators ŽGrasmuck ¨ et al., 1995., particularly in the monitoring of cultural eutrophication. For example, the Environment Agency of England and Wales is monitoring macrophytes upstream and downstream of significant point sources of phosphorus to aid the implementation of the Urban Waste Water Treatment Directive ŽHolmes et al., 1999.. The method developed for this purpose is the Mean Trophic Rank ŽMTR. assessment system ŽHolmes et al., 1999. which uses a 10᎐100 scale based upon scores and cover values of indicator species Žlow MTR values are indicative of eutrophic and polluted reaches .. However, the relationship between aquatic macrophytes and the trophic status of rivers is a complex one not least because of the synergistic effects of a range of environmental variables Žincluding current velocity, substrate, discharge variation, light, overall water quality and nutrients . and the biotic factors of competition and the physical or grazing effects of animals Žsee Demars and Harper, 1998.. Attempts to understand the direct effects of nutrients in lotic systems are further compounded by the fact that submerged, rooted macrophytes can absorb nutrients either by roots or shoots, or by both together in varying proportions ŽBarko et al., 1991..
Results from a range of experiments conducted in laboratories Že.g. Best et al., 1996., lakes Že.g. Carignan and Kalff, 1980., and rivers Že.g. Chambers et al., 1989., which have tried to establish the relative importance of sediments or water as sources of nutrients for aquatic macrophytes, are inconclusive. Denny Ž1980. cautions that it is difficult to compare research findings due to the variety of subject species and techniques employed. Furthermore, it is not possible to apply the results from the more numerous lake studies to running waters: first, because it is likely that the role of macrophytes in nutrient cycling will be more significant in streams where plants are able to colonise much of the channel compared to lakes where light attenuation may restrict macrophytes to the littoral zone; second, in flowing waters the gradients in nutrient concentration will be less predictable than in standing waters ŽChambers et al., 1992.; and third, the greater temporal and spatial heterogeneity of the running water environment ŽDemars and Harper, 1998. will result in areas of both nutrient retention and nutrient throughput. Thus, erosional and depositional processes operating within river channels have the potential to create bed sediments that display a high degree of spatial heterogeneity with fine particles accumulating in areas of reduced flow velocity such as backwaters, sheltered channel margins and within macrophyte stands. In this way, spatial variation in the physical properties of bed sediments can lead to an associated heterogeneity in sediment chemical properties Žcf. Stone and English, 1993.. The influence of flow variability within the channel will enhance the spatial variation in sediment chemistry with turbulence encouraging oxygenation of the sediment᎐water interface and influencing sediment nutrient retention and release ŽBostrom ¨ et al., 1988.. Sand-Jensen Ž1998. demonstrated that macrophyte species with different morphological characteristics influence flow velocity and sediment dynamics to differing degrees. It seems likely, therefore, that the sediments associated with different macrophyte species will possess distinct particle size composition and nutrient concentrations.
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
The factors which appear to be important in determining the role of sediment nutrients in the structure and function of macrophyte communities include plant morphology ŽDenny, 1972., organic matter content and type within the sediment ŽBarko and Smart, 1983., sediment density ŽBarko et al., 1991., water column pH ŽSchuurkes et al., 1986., and the ratio of sediment to water nutrient concentrations or the levels of nutrients in the interstitial compared to the external water. To understand the relationship between macrophytes and nutrients in rivers, and before macrophytes can be used as trophic indicators, research is needed to: establish the spatial and temporal variability of sediment nutrient characteristics in rivers, and the link with water column chemistry; and specify the precise nature of the relationship between sediment nutrients and macrophyte species and communities. The aim of this paper is to present the preliminary results from a study of the spatial variability of sediment nutrient characteristics across 17 rivers in lowland England. The sediment characteristics underlying five aquatic macrophyte species are also examined. The results presented here are part of a continuing and wider investigation into sediment characteristics, macrophyte tolerances to sediment conditions and the influence of larger-scale processes, such as hydrological regime and water chemistry upon sediment nutrient content.
2. Methods Sediment samples were collected from 17 rivers of moderate to high trophic status Ž0.03᎐3.22 mgrl orthophosphate, calculated as the 1-year mean from Environment Agency data. in southern, lowland, England. Eleven of the sampled sites occur on chalk geology Žrivers Allen, Avon, Dun, Frome, Hiz, Itchen, Rhee, Test, Wey, Whitewater and Wylye.; three rivers occur on chalk overlain by boulder clay ŽLoddon, Tove and Waveney.; and three are on clay ŽDove, Eden and Whilton Branch. ŽUK Hydrometric Register, NERC, 1993.. The criteria for site selection were: availability of recent Environment Agency water quality data; wadable reaches to permit sample
105
collection; the presence of at least four submerged, rooted macrophyte species at each site Žfrom a specified list of 30 species which have a widespread distribution in England and usually form distinct stands to ease sampling and delineation of the sample boundary.; and proximity to London to allow speedy analysis of samples following field sampling. At each river site, 20 sediment samples were taken from a 100-m reach Žcomparable to the MTR survey reach, see Holmes et al., 1999. to investigate the degree of spatial variability at each river location. Two of the sediment samples at each site were taken from bare, unvegetated sediment. The remaining samples were taken from underneath stands of the selected plant species with at least two samples per species Žeach from different stands.. Samples were located randomly within macrophyte stands. Sediment was collected using a suction corer of similar construction to the type described by Maitland Ž1969.. A depth of approximately 10 cm was extracted. Samples were placed in a cool box for transport and then stored in a refrigerator Ž; 4⬚C. and in the dark until analyses were performed. Macrophyte surveys were conducted in accordance with the MTR methodology ŽHolmes et al., 1999. to ensure the widest possible application of results in the UK. Macrophyte abundance Žrepresenting percentage cover on a vertical scale. was recorded using the MTR recommended nine-point scale wfrom scale point 1 Ž- 0.1%. to scale point 9 Ž) 75%.x. Information on water column chemistry, geology Žsolid and drift., altitude, distance from source, flow velocity, channel geometry and morphometry, degree of shading and land use was obtained from either archive sources or from field measurement or observation. Each sediment sample was analysed for total phosphorus, inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter content. All chemical analyses were performed on dried samples ground to pass through a 250-mm sieve. Total phosphorus was determined by digestion of sediment with sulfuric acid and sodium sulfate in the presence of a copper sulfate catalyst, followed by soluble reactive phosphorus ŽSRP. determination of the ex-
106
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
tract using the method of Murphy and Riley Ž1962.. Inorganic phosphorus was determined by extracting a 2-g sediment pellet by boiling for 20 min with 1 M HCl ŽAndersen, 1976.. Total nitrogen and organic carbon were simultaneously determined by combustion gas chromatography using a Carlo Erba 1108 Elemental ŽCHN. Analyser Žafter Pella and Columbo, 1973.. Percentage organic matter in each sample was determined as loss on ignition overnight at 450⬚C. Particle size analysis was undertaken to determine the percentage of silt and clay in each sample. Each sediment sample was first passed through a 4-mm sieve to remove the coarsest bed material; a 50᎐100-g sample of this material was passed through a 2-mm sieve to remove gravel; and the remainder was wet-sieved through a 63m sieve to determine the silt᎐clay content. The silt᎐clay fraction was expressed as a percentage of the material finer than 4 mm. The spatial variability of sediment characteristics across the 17 river sites was analysed by plotting boxplot graphs of the six sediment variables Žtotal phosphorus, inorganic phosphorus, total nitrogen, organic carbon, percent silt᎐clay and percent organic matter. by river. A comparison of mean values for the six sediment variables was undertaken with one-way single factor analysis of variance ŽANOVA, Model II. tests. The sediment characteristics underlying five macrophyte species were assessed by plotting boxplot graphs for each species. These five species Ž Elodea nuttallii ŽPlanch.. H. St. John; Myriophyllum spicatum L.; Potamogeton pectinatus L.; Ranunculus penicillatus subsp. pseudofluitans ŽSyme. S.D. Webster; and Sparganium emersum Rehmann. were selected for preliminary analysis because they were the most frequently sampled macrophytes across the 17 rivers and they have a wide distribution in England.
3. Results and discussion Fig. 1a indicates the variability of sediment concentrations of total phosphorus Žtotal P. among and within rivers. Mean total P concentrations for sites ranged from 154 grg ŽAvon. to
2247 grg ŽWey.. The other rivers may be divided into two groups on the basis of their mean total P concentrations with eight rivers having mean concentrations less than 1000 grg and seven rivers with mean concentrations between 1000 and 1500 grg. The different rivers are distinct from one another on the basis of the total P concentrations of their sediments with some rivers having low concentrations with a limited range ŽAllen, Avon, and Whitewater., others spanning a wide range of concentrations ŽLoddon and Waveney., and finally rivers with high total P concentrations across a narrow range ŽTove, Wey and Whilton.. The values obtained for total P concentrations in river sediments are comparable with total P values quoted in the literature, for example 312᎐1776 grg total P in samples from the Gjern and Gelbaek rivers, Denmark, analysed by four different methods ŽSvendsen et al., 1993., and 433᎐627 grg total P from sediments collected in the Camargue and Lake Balaton, France ŽDe Groot and Golterman, 1990.. However, the concentrations are somewhat lower than the highest values measured by Hieltjes and Lijklema Ž1980. and Rose Ž1995.: 746᎐4158 grg total P in Lake Brielle sediments; and 168᎐6158 grg total P in River Welland ŽUK. and River Morava ŽCzech Republic. sediments, respectively. Phosphorus concentrations in river and lake sediments are often high due to the capacity of sediments to bind phosphorus on particle surfaces and as minerals with calcium, iron and aluminium. The sediments may therefore act as a sink Žand subsequently a source if conditions change. for phosphorus within aquatic systems and it has been suggested that as much as 70% of all phosphorus within a system may ultimately end up in the sediments ŽGolterman et al., 1983. and 1 m2 of sediment 10 cm deep may contain 300 times the phosphorus of a 5-m-deep column of overlying water ŽLijklema, 1998.. Fig. 1b suggests that inorganic phosphorus Žinorganic P. concentrations follow a similar pattern to total P concentrations in the rivers reflecting the fact that inorganic P represents a proportion of total P. As most of the rivers sampled are on neutral or calcareous geology, it is likely that much of this inorganic P is associated with cal-
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
107
Fig. 1. Boxplots of sediment characteristics plotted by river. Note: Ž1. values were obtained from analyses of 20 sediment samples per river; Ž2. the 17 rivers are: ALL ŽAllen.; AVO ŽAvon.; DOV ŽDove.; DUN ŽDun.; EDE ŽEden.; FRO ŽFrome.; HIZ ŽHiz.; ITC ŽItchen .; LOD ŽLoddon.; RHE ŽRhee.; TES ŽTest.; TOV ŽTove.; WAV ŽWaveney.; WEY ŽWey.; WHI ŽWhilton Branch.; WWA ŽWhitewater.; and WYL ŽWylye..
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
108
cium rather than iron or aluminium as in acidic sediments ŽHesse, 1973.. The highest and lowest mean inorganic P concentrations 750 grg and 39 grg are for the Wey and Avon, respectively. Particular rivers display different ranges for inorganic P compared to total P, for example the Tove sediments have high total P concentrations with a small range but relatively low inorganic P concentrations with lots of within site variability. Similarly, the Wey also has much greater variability in inorganic P concentrations than total P concentrations. Fig. 1c,d suggest that concentrations of total nitrogen Žtotal N. and organic carbon Žorganic C. in the river sediments are closely related. Both are products of the breakdown of organic matter and thus concentrations will be mediated by microbial activity in the sediment. Mean total N contents range from 0.02 to 0.52% for the Eden and Waveney respectively, and almost all samples from rivers other than the Waveney were found to have total N contents less than 0.5%. The nitrogen concentrations are of similar magnitude to those measured in 31 rivers in north-east England Žmean 0.06%, range - 0.001᎐0.51% . ŽGarcıa-Ruiz et al., 1998.. Nitrogen is evidently a ´ very small component of these river sediments; sediments are depleted of nitrogen more rapidly than phosphorus, due to the smaller exchangeable pools of nitrogen buffering the plant-available nitrogen in the interstitial water ŽBarko et al., 1991.. Minimum and maximum mean organic C concentrations are also found in the Eden Ž0.60%. and Waveney Ž9.07%., respectively.
Whilst organic C is present in the sediments in quantities often an order of magnitude greater than total N proportions, it still represents a fairly small proportion of the sediment. Variability of these two sediment nutrients is greatest in the River Waveney with other rivers having relatively consistent values within a site. Mean percent silt᎐clay values vary from 4% in the Avon to 45% in the Waveney. However, 12 of the 17 rivers have mean silt᎐clay percentages of between 20 and 30%. Fig. 1e indicates that many of the rivers have similar ranges of silt᎐clay percentages suggesting similar degrees of heterogeneity although the rivers Avon and Waveney are distinct in having the smallest and largest ranges of silt᎐clay contents, respectively. The sandy nature of the Avon sediments may explain the low phosphorus concentrations observed at this site ŽFig. 1a,b. as sediment phosphorus concentrations and the proportion of fine particles are often closely related Žcf. Chambers et al., 1992; Stone and English, 1993.. Mean organic matter content of the river sediments varies from 2 to 21% in the Hiz and Waveney, respectively ŽFig. 1f.. All rivers excluding the Waveney had organic matter contents of 10% or less suggesting that in many cases organic matter constitutes only a small component of sediment by weight. The sediments of the Waveney are distinct from those of the other rivers on the basis of all the measured variables ŽFig. 1a᎐f.. The Waveney was observed to have little flow and almost total cover of macrophytes at the time of sampling. It is proposed that the
Table 1 Pearson’s product moment correlation coefficients for sediment parameters across all 17 rivers a
TP IP TN OC OM SC a
TP
IP
TN
OC
OM
SC
1 U 0.858 0.047 0.089 U 0.184 U 0.179
1 y0.038 0.023 0.090 0.021
1 U 0.941 U 0.725 U 0.590
1 U 0.820 U 0.576
1 U 0.574
1
Abbre¨ iations: TP, total phosphorus; IP, inorganic phosphorus; TN, total nitrogen; OC, organic carbon; OM, organic matter SC, siltrclay. U Correlation is significant at the 0.01 level Žtwo tailed..
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
absence of flow and abundant macrophyte growth in the river resulted in the build-up and retention of organic matter Žcf. macrophyte patches described by Sand-Jensen, 1998. and in situ decomposition of macrophyte material leading to elevated concentrations of total N and organic C and a high proportion of fines in the sediments. The spatial variability of sediment characteristics displayed in Fig. 1a᎐f was investigated by performing one-way single factor ANOVA, Model II tests. Results of these tests indicate that for each of the variables Žtotal P, inorganic P, total N, organic C, % silt-clay, % organic matter. the means were significantly different among the 17 rivers Ž Ps 0.000.. Table 1 displays Pearson’s product moment correlation coefficients for the six sediment variables. The coefficients confirm the strong linear association between organic C, total N and organic matter content of the sediments. This relationship reflects the common origin of carbon and nitrogen which are both released from organic material upon decomposition. Close relationships between organic C and total nitrogen have also been noted in Russian lake sediments ŽMartinova, 1993.. Fig. 2a displays the total P concentrations for the sediments within which the five macrophyte species were rooted. E. nuttallii is associated with the greatest range of total P concentrations with a range of almost 2500 grg. R. penicillatus subsp. pseudofluitans and S. emersum have similar ranges if the values denoted as outliers are included. P. pectinatus is normally associated with enriched waters ŽPreston and Croft, 1997.; in this study, however, it is associated with only moderate total P concentrations relative to the other four species. M. spicatum is found on sediments with relatively low total P concentrations possibly reflecting its preference for sandy sediments ŽFig. 2e.. Fig. 2b indicates that the species are distributed with respect to inorganic P in a similar manner to total P concentration. Total N contents are low for sediments under all species and if extreme values are excluded the ranges are all quite small ŽFig. 2c.. This may be indicative of rapid uptake of nitrogen by the plants and accumulation in plant tissue. Howard-
109
Williams and Downes Ž1984. Žas cited by Howard-Williams, 1985. provide evidence of rapid nitrogen uptake calculating plant uptake at a rate of 900᎐1500 mg Nrm2 per day in a macrophyte dominated New Zealand stream. Organic C contents display a similar range and pattern to the total N contents, also displayed in the river plots ŽFig. 1c,d.. Levels of organic matter, nitrogen and carbon may be a direct result of retention by macrophyte stands; organic matter and nitrogen concentrations of sediments of the River Susa ŽDenmark. were close to zero during the winter but as flows were reduced and macrophyte cover increased during the summer a 1-cm-thick layer of organic matter accumulated on the sediment surface ŽSand-Jensen et al., 1989.. The species discussed are all associated with sediments with mean% silt᎐clay contents ranging from 13.73 to 26.39% ŽFig. 2e.. M. spicatum is found on the sandiest sediments and across the most restricted range of particle size types. This may explain the low inorganic P nature of M. spicatum sediments ŽFig. 2b., as inorganic P will bind to the surfaces of finer sediments. Fig. 2e shows that P. pectinatus and R. penicillatus subsp. pseudofluitans grow on similar sized sediments reflecting their similar ecological niche and explaining why P. pectinatus may replace R. penicillatus subsp. pseudofluitans in eutrophic or sluggish waters ŽPreston and Croft, 1997.. S. emersum is also associated with a similar range of silt᎐clay contents although it roots more deeply and forms a more open canopy than R. penicillatus subsp. pseudofluitans and P. pectinatus and is likely to have less of an effect on sediment dynamics. Organic matter contents are fairly low Ž- 10% of sediment by weight. for most samples underlying the five species. This preliminary analysis of the sediments of 17 lowland eutrophic rivers indicates that sediment characteristics Žtotal and inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter. are highly variable even at the scale of a 100-m river reach. However, the ANOVA tests demonstrate there is even greater variability between rivers despite the limited geographical and trophic range. Clearly, the role of other factors, such as water chemistry, hydrologi-
110
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
Fig. 2. Boxplots of sediment characteristics plotted by macrophyte species. Note: Ž1. please refer to the key in Fig. 1 for explanation of boxplots; Ž2. the five aquatic macrophyte species are: Elod nut Ž Elodea nuttallii ŽPlanch.. H. St. John.; Myri spi Ž Myriophyllum spicatum L..; Pota pec Ž Potamogeton pectinatus L..; Ranu pen Ž Ranunculus penicillatus subsp. pseudofluitans ŽSyme. S.D. Webster.; and Spar eme Ž Sparganium emersum Rehmann..
cal regime, catchment geology and landuse, will also be important in determining sediment character and spatial variability. This aspect is cur-
rently being investigated through multivariate analysis. The data presented in this paper give some indication of the sediment nutrient charac-
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
teristics associated with the five macrophyte species common in the UK but it is difficult to ascribe sediment preferences for particular species without further investigation. In particular, more research is needed to establish the precise nature of the relationships between the physical and chemical properties of sediments and macrophyte species and communities for a wider range of river types, including investigations into the temporal and spatial variability of these relationships.
Acknowledgements The authors gratefully acknowledge the financial support of the Environment Agency, and Queen Mary & Westfield College through the provision of a Drapers Scholarship to SJC and a Social Science Research grant to GW. The research could not have been undertaken without the advice and support of Nigel Holmes and Mark Everard. References Andersen JM. An ignition method for determination of total phosphorus in lake sediments. Water Res 1976;10:329᎐331. Barko JW, Smart RM. Effects of organic matter additions to sediment on the growth of aquatic plants. J Ecol 1983; 71:161᎐175. Barko JW, Gunnison D, Carpenter SR. Sediment interactions with submerged macrophyte growth and community dynamics. Aquatic Botany 1991;41:41᎐65. Best EPH, Woltman H, Jacobs FHH. Sediment-related growth limitation of Elodea nuttallii as indicated by a fertilization experiment. Freshwater Biol 1996;36:33᎐44. Bostrom ¨ B, Andersen JM, Fleischer S, Jansson M. Exchange of phosphorus across the sediment᎐water interface. Hydrobiologia 1988;170:229᎐244. Canfield DE, Langeland KA, Maceina MJ, Haller WT, Shireman JV. Trophic state classification of lakes with aquatic macrophytes. Can J Fisheries Aquatic Sci 1983;40: 1713᎐1718. Carignan R, Kalff J. Phosphorus sources for aquatic weeds: water or sediments? Science 1980;207:987᎐989. Chambers PA, Prepas EE, Bothwell ML, Hamilton HR. Roots versus shoots in nutrient uptake by aquatic macrophytes in flowing waters. Can J Fisheries Aquatic Sci 1989;46: 435᎐439. Chambers PA, Prepas EE, Gibson K. Temporal and spatial dynamics in riverbed chemistry: the influence of flow and
111
sediment composition. Can J Fisheries Aquatic Sci 1992;49:2128᎐2140. De Groot CJ, Golterman HL. Sequential fractionation of sediment phosphate. Hydrobiologia 1990;192:143᎐148. Demars BOL, Harper DM. The aquatic macrophytes of an English lowland river system: assessing response to nutrient enrichment. Hydrobiologia 1998;384:75᎐88. Denny P. Sites of nutrient absorption in aquatic macrophytes. J Ecol 1972;60:819᎐829. Denny P. Solute movement in submerged angiosperms. Biol Rev 1980;55:65᎐92. Garcıa-Ruiz R, Pattinson SN, Whitton BA. Denitrification in ´ river sediments: relationship between process rate and properties of water and sediment. Freshwater Biol 1998;39:467᎐476. Grasmuck L, Muller S. Assessment of the ¨ N, Haury J, Leglize ´ bio-indicator capacity of aquatic plants using multivariate analysis. Hydrobiologia 1995;300r301:115᎐122. Golterman HL, Sly PG, Thomas RL. Study of the relationship between water quality and sediment transport. Technical Papers in Hydrology 26 ŽA contribution to the International Hydrological Programme., UNESCO, Paris, 1983, 222 pp. Hesse PR. Phosphorus in lake sediments. In: Griffith EJ, Beeton A, Spencer JM, Mitchell DT, editors. Environmental phosphorus handbook. New York: John Wiley and Sons, 1973:573᎐583. Hieltjes AHM, Lijklema L. Fractionation of inorganic phosphates in calcareous sediments. J Environ Qual 1980; 9:405᎐407. Holmes NTH, Newman JR, Chadd S, Rouen KJ, Saint L, Dawson FH. Mean trophic rank: a user’s manual. R& D Technical Report E38. Bristol, UK: Environment Agency, 1999. Howard-Williams C. Cycling and retention of nitrogen and phosphorus in wetlands: a theoretical and applied perspective. Freshwater Biol 1985;15:391᎐431. Lijklema L. Dimensions and scales. Water Sci Technol 1998;37:1᎐7. Maitland PS. A simple corer for sampling sand and finer sediments in shallow water. Limnol Oceanogr 1969;14: 151᎐156. Martinova MV. Nitrogen and phosphor compounds in bottom sediments: mechanisms of accumulation, transformation and release. Hydrobiologia 1993;252:1᎐22. Murphy J, Riley JP. A modified single solution method for the determination of phosphate in natural waters. Anal Chim Acta 1962;27:31᎐36. Natural Environment Research Council ŽNERC.. Hydrological data United Kingdom: hydrometric register and statistics 1986᎐90. Wallingford: Institute of Hydrology, 1993. 179 pp.. Pella E, Columbo B. The study of carbon hydrogen and nitrogen by combustion-gas chromatography. Mikrochem Acta 1973;5:697᎐719. Preston CD, Croft JM. Aquatic plants in Britain and Ireland. Colchester, England: Harley Books, 1997:365.
112
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
Rose M. The role of sediment phosphorus concentration in controlling the growth of macrophytes in moving waters. Unpublished MSc Thesis, Dept. of Zoology, University of Leicester, 1995, 41 pp. Sand-Jensen K. Macrophytes as biological engineers in the ecology of Danish streams. In: Sand-Jensen K, Pedersen O, editors. Freshwater biology: priorities and development in Danish research, The Freshwater Biological Laboratory. Copenhagen: University of Copenhagen and G.E.C. Gad Publishers Ltd, 1997:74᎐101. Sand-Jensen K. Influence of submerged macrophytes on sediment composition and near-bed flow in lowland streams. Freshwater Biol 1998;39:663᎐679.
Sand-Jensen K, Jeppesen E, Nielsen K et al. Growth of macrophytes and ecosystem consequences in a lowland Danish stream. Freshwater Biol 1989;22:15᎐32. Schuurkes JA, Kok CJ, Den Hartog C. Ammonium and nitrate uptake by aquatic plants from poorly buffered and acidified waters. Aquatic Botany 1986;24:131᎐146. Stone M, English MC. Geochemical composition, phosphorus speciation and mass transport of fine-grained sediment in two Lake Erie tributaries. Hydrobiologia 1993;253:17᎐29. Svendsen LM, Rebsdorf A, Nornberg P. Comparison of methods for analysis of organic and inorganic phosphorus in river sediment. Water Res 1993;27:77᎐83.
Sediment nutrient characteristics and aquatic macrophytes in lowland English rivers Stewart J. Clarke, Geraldene WhartonU Department of Geography, Queen Mary and Westfield College, Uni¨ ersity of London, London E1 4NS, UK Received 17 September 1999; accepted 28 June 2000
Abstract Aquatic macrophytes play an important role in the nutrient dynamics of streams. As a result, there is much interest in their use as trophic indicators. However, the relationship between aquatic macrophytes and the trophic status of rivers is a complex one, partly because of the effects of a wide range of environmental variables and partly because submerged, rooted macrophytes can absorb nutrients from the river sediments andror the water column. Experiments which have tried to establish the relative importance of sediments or water as sources of nutrients are inconclusive and further work is needed to establish how sediment nutrient characteristics vary within and among rivers Žspatially and temporally. and the inter-relationships between sediment nutrients, water column chemistry and macrophytes. This paper presents the initial findings from a study of 17 lowland rivers in southern England which is exploring the spatial variability of sediment characteristics Žtotal and inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter content. and the relationship with aquatic macrophytes. The preliminary analysis indicates that although sediment characteristics are highly variable within 100-m river reaches, the variability across the 17 rivers is even greater; this is despite the limited geographic and trophic range of the study sites. The results presented in this paper also give some indication of the sediment characteristics associated with five macrophyte species but it is too early to ascribe sediment preferences for particular species. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Aquatic macrophytes; Sediments; Nutrients; Trophic indicator
U
Corresponding author. Tel.: q44-20-7882-5436; fax: q44-20-8981-6276. E-mail address: [email protected] ŽG. Wharton.. 0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 7 5 4 - 3
104
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
1. Introduction Shallow, low-energy, temperate streams, of the type found in lowland England, often have low flow velocities and fine, nutrient-rich sediments that result in abundant macrophyte growth ŽDemars and Harper, 1998; cf. Danish streams as described by Sand-Jensen, 1997.. Under these conditions, macrophytes Ždescribed as ‘biological engineers’ by Sand-Jensen, 1997. play a key role in nutrient dynamics by: trapping organic matter and debris ŽSand-Jensen, 1998.; acting as temporary sinks for nutrients during periods of growth Žcf. Canfield et al., 1983.; and functioning as a link between the sediment and water column through the release of nutrients, for example from senescent and decaying stems and leaves ŽCarignan and Kalff, 1980. or from the sediment through the creation of anoxic sediments within dense plant stands. As a result, there is much interest in aquatic macrophytes as trophic indicators ŽGrasmuck ¨ et al., 1995., particularly in the monitoring of cultural eutrophication. For example, the Environment Agency of England and Wales is monitoring macrophytes upstream and downstream of significant point sources of phosphorus to aid the implementation of the Urban Waste Water Treatment Directive ŽHolmes et al., 1999.. The method developed for this purpose is the Mean Trophic Rank ŽMTR. assessment system ŽHolmes et al., 1999. which uses a 10᎐100 scale based upon scores and cover values of indicator species Žlow MTR values are indicative of eutrophic and polluted reaches .. However, the relationship between aquatic macrophytes and the trophic status of rivers is a complex one not least because of the synergistic effects of a range of environmental variables Žincluding current velocity, substrate, discharge variation, light, overall water quality and nutrients . and the biotic factors of competition and the physical or grazing effects of animals Žsee Demars and Harper, 1998.. Attempts to understand the direct effects of nutrients in lotic systems are further compounded by the fact that submerged, rooted macrophytes can absorb nutrients either by roots or shoots, or by both together in varying proportions ŽBarko et al., 1991..
Results from a range of experiments conducted in laboratories Že.g. Best et al., 1996., lakes Že.g. Carignan and Kalff, 1980., and rivers Že.g. Chambers et al., 1989., which have tried to establish the relative importance of sediments or water as sources of nutrients for aquatic macrophytes, are inconclusive. Denny Ž1980. cautions that it is difficult to compare research findings due to the variety of subject species and techniques employed. Furthermore, it is not possible to apply the results from the more numerous lake studies to running waters: first, because it is likely that the role of macrophytes in nutrient cycling will be more significant in streams where plants are able to colonise much of the channel compared to lakes where light attenuation may restrict macrophytes to the littoral zone; second, in flowing waters the gradients in nutrient concentration will be less predictable than in standing waters ŽChambers et al., 1992.; and third, the greater temporal and spatial heterogeneity of the running water environment ŽDemars and Harper, 1998. will result in areas of both nutrient retention and nutrient throughput. Thus, erosional and depositional processes operating within river channels have the potential to create bed sediments that display a high degree of spatial heterogeneity with fine particles accumulating in areas of reduced flow velocity such as backwaters, sheltered channel margins and within macrophyte stands. In this way, spatial variation in the physical properties of bed sediments can lead to an associated heterogeneity in sediment chemical properties Žcf. Stone and English, 1993.. The influence of flow variability within the channel will enhance the spatial variation in sediment chemistry with turbulence encouraging oxygenation of the sediment᎐water interface and influencing sediment nutrient retention and release ŽBostrom ¨ et al., 1988.. Sand-Jensen Ž1998. demonstrated that macrophyte species with different morphological characteristics influence flow velocity and sediment dynamics to differing degrees. It seems likely, therefore, that the sediments associated with different macrophyte species will possess distinct particle size composition and nutrient concentrations.
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
The factors which appear to be important in determining the role of sediment nutrients in the structure and function of macrophyte communities include plant morphology ŽDenny, 1972., organic matter content and type within the sediment ŽBarko and Smart, 1983., sediment density ŽBarko et al., 1991., water column pH ŽSchuurkes et al., 1986., and the ratio of sediment to water nutrient concentrations or the levels of nutrients in the interstitial compared to the external water. To understand the relationship between macrophytes and nutrients in rivers, and before macrophytes can be used as trophic indicators, research is needed to: establish the spatial and temporal variability of sediment nutrient characteristics in rivers, and the link with water column chemistry; and specify the precise nature of the relationship between sediment nutrients and macrophyte species and communities. The aim of this paper is to present the preliminary results from a study of the spatial variability of sediment nutrient characteristics across 17 rivers in lowland England. The sediment characteristics underlying five aquatic macrophyte species are also examined. The results presented here are part of a continuing and wider investigation into sediment characteristics, macrophyte tolerances to sediment conditions and the influence of larger-scale processes, such as hydrological regime and water chemistry upon sediment nutrient content.
2. Methods Sediment samples were collected from 17 rivers of moderate to high trophic status Ž0.03᎐3.22 mgrl orthophosphate, calculated as the 1-year mean from Environment Agency data. in southern, lowland, England. Eleven of the sampled sites occur on chalk geology Žrivers Allen, Avon, Dun, Frome, Hiz, Itchen, Rhee, Test, Wey, Whitewater and Wylye.; three rivers occur on chalk overlain by boulder clay ŽLoddon, Tove and Waveney.; and three are on clay ŽDove, Eden and Whilton Branch. ŽUK Hydrometric Register, NERC, 1993.. The criteria for site selection were: availability of recent Environment Agency water quality data; wadable reaches to permit sample
105
collection; the presence of at least four submerged, rooted macrophyte species at each site Žfrom a specified list of 30 species which have a widespread distribution in England and usually form distinct stands to ease sampling and delineation of the sample boundary.; and proximity to London to allow speedy analysis of samples following field sampling. At each river site, 20 sediment samples were taken from a 100-m reach Žcomparable to the MTR survey reach, see Holmes et al., 1999. to investigate the degree of spatial variability at each river location. Two of the sediment samples at each site were taken from bare, unvegetated sediment. The remaining samples were taken from underneath stands of the selected plant species with at least two samples per species Žeach from different stands.. Samples were located randomly within macrophyte stands. Sediment was collected using a suction corer of similar construction to the type described by Maitland Ž1969.. A depth of approximately 10 cm was extracted. Samples were placed in a cool box for transport and then stored in a refrigerator Ž; 4⬚C. and in the dark until analyses were performed. Macrophyte surveys were conducted in accordance with the MTR methodology ŽHolmes et al., 1999. to ensure the widest possible application of results in the UK. Macrophyte abundance Žrepresenting percentage cover on a vertical scale. was recorded using the MTR recommended nine-point scale wfrom scale point 1 Ž- 0.1%. to scale point 9 Ž) 75%.x. Information on water column chemistry, geology Žsolid and drift., altitude, distance from source, flow velocity, channel geometry and morphometry, degree of shading and land use was obtained from either archive sources or from field measurement or observation. Each sediment sample was analysed for total phosphorus, inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter content. All chemical analyses were performed on dried samples ground to pass through a 250-mm sieve. Total phosphorus was determined by digestion of sediment with sulfuric acid and sodium sulfate in the presence of a copper sulfate catalyst, followed by soluble reactive phosphorus ŽSRP. determination of the ex-
106
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
tract using the method of Murphy and Riley Ž1962.. Inorganic phosphorus was determined by extracting a 2-g sediment pellet by boiling for 20 min with 1 M HCl ŽAndersen, 1976.. Total nitrogen and organic carbon were simultaneously determined by combustion gas chromatography using a Carlo Erba 1108 Elemental ŽCHN. Analyser Žafter Pella and Columbo, 1973.. Percentage organic matter in each sample was determined as loss on ignition overnight at 450⬚C. Particle size analysis was undertaken to determine the percentage of silt and clay in each sample. Each sediment sample was first passed through a 4-mm sieve to remove the coarsest bed material; a 50᎐100-g sample of this material was passed through a 2-mm sieve to remove gravel; and the remainder was wet-sieved through a 63m sieve to determine the silt᎐clay content. The silt᎐clay fraction was expressed as a percentage of the material finer than 4 mm. The spatial variability of sediment characteristics across the 17 river sites was analysed by plotting boxplot graphs of the six sediment variables Žtotal phosphorus, inorganic phosphorus, total nitrogen, organic carbon, percent silt᎐clay and percent organic matter. by river. A comparison of mean values for the six sediment variables was undertaken with one-way single factor analysis of variance ŽANOVA, Model II. tests. The sediment characteristics underlying five macrophyte species were assessed by plotting boxplot graphs for each species. These five species Ž Elodea nuttallii ŽPlanch.. H. St. John; Myriophyllum spicatum L.; Potamogeton pectinatus L.; Ranunculus penicillatus subsp. pseudofluitans ŽSyme. S.D. Webster; and Sparganium emersum Rehmann. were selected for preliminary analysis because they were the most frequently sampled macrophytes across the 17 rivers and they have a wide distribution in England.
3. Results and discussion Fig. 1a indicates the variability of sediment concentrations of total phosphorus Žtotal P. among and within rivers. Mean total P concentrations for sites ranged from 154 grg ŽAvon. to
2247 grg ŽWey.. The other rivers may be divided into two groups on the basis of their mean total P concentrations with eight rivers having mean concentrations less than 1000 grg and seven rivers with mean concentrations between 1000 and 1500 grg. The different rivers are distinct from one another on the basis of the total P concentrations of their sediments with some rivers having low concentrations with a limited range ŽAllen, Avon, and Whitewater., others spanning a wide range of concentrations ŽLoddon and Waveney., and finally rivers with high total P concentrations across a narrow range ŽTove, Wey and Whilton.. The values obtained for total P concentrations in river sediments are comparable with total P values quoted in the literature, for example 312᎐1776 grg total P in samples from the Gjern and Gelbaek rivers, Denmark, analysed by four different methods ŽSvendsen et al., 1993., and 433᎐627 grg total P from sediments collected in the Camargue and Lake Balaton, France ŽDe Groot and Golterman, 1990.. However, the concentrations are somewhat lower than the highest values measured by Hieltjes and Lijklema Ž1980. and Rose Ž1995.: 746᎐4158 grg total P in Lake Brielle sediments; and 168᎐6158 grg total P in River Welland ŽUK. and River Morava ŽCzech Republic. sediments, respectively. Phosphorus concentrations in river and lake sediments are often high due to the capacity of sediments to bind phosphorus on particle surfaces and as minerals with calcium, iron and aluminium. The sediments may therefore act as a sink Žand subsequently a source if conditions change. for phosphorus within aquatic systems and it has been suggested that as much as 70% of all phosphorus within a system may ultimately end up in the sediments ŽGolterman et al., 1983. and 1 m2 of sediment 10 cm deep may contain 300 times the phosphorus of a 5-m-deep column of overlying water ŽLijklema, 1998.. Fig. 1b suggests that inorganic phosphorus Žinorganic P. concentrations follow a similar pattern to total P concentrations in the rivers reflecting the fact that inorganic P represents a proportion of total P. As most of the rivers sampled are on neutral or calcareous geology, it is likely that much of this inorganic P is associated with cal-
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
107
Fig. 1. Boxplots of sediment characteristics plotted by river. Note: Ž1. values were obtained from analyses of 20 sediment samples per river; Ž2. the 17 rivers are: ALL ŽAllen.; AVO ŽAvon.; DOV ŽDove.; DUN ŽDun.; EDE ŽEden.; FRO ŽFrome.; HIZ ŽHiz.; ITC ŽItchen .; LOD ŽLoddon.; RHE ŽRhee.; TES ŽTest.; TOV ŽTove.; WAV ŽWaveney.; WEY ŽWey.; WHI ŽWhilton Branch.; WWA ŽWhitewater.; and WYL ŽWylye..
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
108
cium rather than iron or aluminium as in acidic sediments ŽHesse, 1973.. The highest and lowest mean inorganic P concentrations 750 grg and 39 grg are for the Wey and Avon, respectively. Particular rivers display different ranges for inorganic P compared to total P, for example the Tove sediments have high total P concentrations with a small range but relatively low inorganic P concentrations with lots of within site variability. Similarly, the Wey also has much greater variability in inorganic P concentrations than total P concentrations. Fig. 1c,d suggest that concentrations of total nitrogen Žtotal N. and organic carbon Žorganic C. in the river sediments are closely related. Both are products of the breakdown of organic matter and thus concentrations will be mediated by microbial activity in the sediment. Mean total N contents range from 0.02 to 0.52% for the Eden and Waveney respectively, and almost all samples from rivers other than the Waveney were found to have total N contents less than 0.5%. The nitrogen concentrations are of similar magnitude to those measured in 31 rivers in north-east England Žmean 0.06%, range - 0.001᎐0.51% . ŽGarcıa-Ruiz et al., 1998.. Nitrogen is evidently a ´ very small component of these river sediments; sediments are depleted of nitrogen more rapidly than phosphorus, due to the smaller exchangeable pools of nitrogen buffering the plant-available nitrogen in the interstitial water ŽBarko et al., 1991.. Minimum and maximum mean organic C concentrations are also found in the Eden Ž0.60%. and Waveney Ž9.07%., respectively.
Whilst organic C is present in the sediments in quantities often an order of magnitude greater than total N proportions, it still represents a fairly small proportion of the sediment. Variability of these two sediment nutrients is greatest in the River Waveney with other rivers having relatively consistent values within a site. Mean percent silt᎐clay values vary from 4% in the Avon to 45% in the Waveney. However, 12 of the 17 rivers have mean silt᎐clay percentages of between 20 and 30%. Fig. 1e indicates that many of the rivers have similar ranges of silt᎐clay percentages suggesting similar degrees of heterogeneity although the rivers Avon and Waveney are distinct in having the smallest and largest ranges of silt᎐clay contents, respectively. The sandy nature of the Avon sediments may explain the low phosphorus concentrations observed at this site ŽFig. 1a,b. as sediment phosphorus concentrations and the proportion of fine particles are often closely related Žcf. Chambers et al., 1992; Stone and English, 1993.. Mean organic matter content of the river sediments varies from 2 to 21% in the Hiz and Waveney, respectively ŽFig. 1f.. All rivers excluding the Waveney had organic matter contents of 10% or less suggesting that in many cases organic matter constitutes only a small component of sediment by weight. The sediments of the Waveney are distinct from those of the other rivers on the basis of all the measured variables ŽFig. 1a᎐f.. The Waveney was observed to have little flow and almost total cover of macrophytes at the time of sampling. It is proposed that the
Table 1 Pearson’s product moment correlation coefficients for sediment parameters across all 17 rivers a
TP IP TN OC OM SC a
TP
IP
TN
OC
OM
SC
1 U 0.858 0.047 0.089 U 0.184 U 0.179
1 y0.038 0.023 0.090 0.021
1 U 0.941 U 0.725 U 0.590
1 U 0.820 U 0.576
1 U 0.574
1
Abbre¨ iations: TP, total phosphorus; IP, inorganic phosphorus; TN, total nitrogen; OC, organic carbon; OM, organic matter SC, siltrclay. U Correlation is significant at the 0.01 level Žtwo tailed..
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
absence of flow and abundant macrophyte growth in the river resulted in the build-up and retention of organic matter Žcf. macrophyte patches described by Sand-Jensen, 1998. and in situ decomposition of macrophyte material leading to elevated concentrations of total N and organic C and a high proportion of fines in the sediments. The spatial variability of sediment characteristics displayed in Fig. 1a᎐f was investigated by performing one-way single factor ANOVA, Model II tests. Results of these tests indicate that for each of the variables Žtotal P, inorganic P, total N, organic C, % silt-clay, % organic matter. the means were significantly different among the 17 rivers Ž Ps 0.000.. Table 1 displays Pearson’s product moment correlation coefficients for the six sediment variables. The coefficients confirm the strong linear association between organic C, total N and organic matter content of the sediments. This relationship reflects the common origin of carbon and nitrogen which are both released from organic material upon decomposition. Close relationships between organic C and total nitrogen have also been noted in Russian lake sediments ŽMartinova, 1993.. Fig. 2a displays the total P concentrations for the sediments within which the five macrophyte species were rooted. E. nuttallii is associated with the greatest range of total P concentrations with a range of almost 2500 grg. R. penicillatus subsp. pseudofluitans and S. emersum have similar ranges if the values denoted as outliers are included. P. pectinatus is normally associated with enriched waters ŽPreston and Croft, 1997.; in this study, however, it is associated with only moderate total P concentrations relative to the other four species. M. spicatum is found on sediments with relatively low total P concentrations possibly reflecting its preference for sandy sediments ŽFig. 2e.. Fig. 2b indicates that the species are distributed with respect to inorganic P in a similar manner to total P concentration. Total N contents are low for sediments under all species and if extreme values are excluded the ranges are all quite small ŽFig. 2c.. This may be indicative of rapid uptake of nitrogen by the plants and accumulation in plant tissue. Howard-
109
Williams and Downes Ž1984. Žas cited by Howard-Williams, 1985. provide evidence of rapid nitrogen uptake calculating plant uptake at a rate of 900᎐1500 mg Nrm2 per day in a macrophyte dominated New Zealand stream. Organic C contents display a similar range and pattern to the total N contents, also displayed in the river plots ŽFig. 1c,d.. Levels of organic matter, nitrogen and carbon may be a direct result of retention by macrophyte stands; organic matter and nitrogen concentrations of sediments of the River Susa ŽDenmark. were close to zero during the winter but as flows were reduced and macrophyte cover increased during the summer a 1-cm-thick layer of organic matter accumulated on the sediment surface ŽSand-Jensen et al., 1989.. The species discussed are all associated with sediments with mean% silt᎐clay contents ranging from 13.73 to 26.39% ŽFig. 2e.. M. spicatum is found on the sandiest sediments and across the most restricted range of particle size types. This may explain the low inorganic P nature of M. spicatum sediments ŽFig. 2b., as inorganic P will bind to the surfaces of finer sediments. Fig. 2e shows that P. pectinatus and R. penicillatus subsp. pseudofluitans grow on similar sized sediments reflecting their similar ecological niche and explaining why P. pectinatus may replace R. penicillatus subsp. pseudofluitans in eutrophic or sluggish waters ŽPreston and Croft, 1997.. S. emersum is also associated with a similar range of silt᎐clay contents although it roots more deeply and forms a more open canopy than R. penicillatus subsp. pseudofluitans and P. pectinatus and is likely to have less of an effect on sediment dynamics. Organic matter contents are fairly low Ž- 10% of sediment by weight. for most samples underlying the five species. This preliminary analysis of the sediments of 17 lowland eutrophic rivers indicates that sediment characteristics Žtotal and inorganic phosphorus, total nitrogen, organic carbon, silt᎐clay fraction and organic matter. are highly variable even at the scale of a 100-m river reach. However, the ANOVA tests demonstrate there is even greater variability between rivers despite the limited geographical and trophic range. Clearly, the role of other factors, such as water chemistry, hydrologi-
110
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
Fig. 2. Boxplots of sediment characteristics plotted by macrophyte species. Note: Ž1. please refer to the key in Fig. 1 for explanation of boxplots; Ž2. the five aquatic macrophyte species are: Elod nut Ž Elodea nuttallii ŽPlanch.. H. St. John.; Myri spi Ž Myriophyllum spicatum L..; Pota pec Ž Potamogeton pectinatus L..; Ranu pen Ž Ranunculus penicillatus subsp. pseudofluitans ŽSyme. S.D. Webster.; and Spar eme Ž Sparganium emersum Rehmann..
cal regime, catchment geology and landuse, will also be important in determining sediment character and spatial variability. This aspect is cur-
rently being investigated through multivariate analysis. The data presented in this paper give some indication of the sediment nutrient charac-
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
teristics associated with the five macrophyte species common in the UK but it is difficult to ascribe sediment preferences for particular species without further investigation. In particular, more research is needed to establish the precise nature of the relationships between the physical and chemical properties of sediments and macrophyte species and communities for a wider range of river types, including investigations into the temporal and spatial variability of these relationships.
Acknowledgements The authors gratefully acknowledge the financial support of the Environment Agency, and Queen Mary & Westfield College through the provision of a Drapers Scholarship to SJC and a Social Science Research grant to GW. The research could not have been undertaken without the advice and support of Nigel Holmes and Mark Everard. References Andersen JM. An ignition method for determination of total phosphorus in lake sediments. Water Res 1976;10:329᎐331. Barko JW, Smart RM. Effects of organic matter additions to sediment on the growth of aquatic plants. J Ecol 1983; 71:161᎐175. Barko JW, Gunnison D, Carpenter SR. Sediment interactions with submerged macrophyte growth and community dynamics. Aquatic Botany 1991;41:41᎐65. Best EPH, Woltman H, Jacobs FHH. Sediment-related growth limitation of Elodea nuttallii as indicated by a fertilization experiment. Freshwater Biol 1996;36:33᎐44. Bostrom ¨ B, Andersen JM, Fleischer S, Jansson M. Exchange of phosphorus across the sediment᎐water interface. Hydrobiologia 1988;170:229᎐244. Canfield DE, Langeland KA, Maceina MJ, Haller WT, Shireman JV. Trophic state classification of lakes with aquatic macrophytes. Can J Fisheries Aquatic Sci 1983;40: 1713᎐1718. Carignan R, Kalff J. Phosphorus sources for aquatic weeds: water or sediments? Science 1980;207:987᎐989. Chambers PA, Prepas EE, Bothwell ML, Hamilton HR. Roots versus shoots in nutrient uptake by aquatic macrophytes in flowing waters. Can J Fisheries Aquatic Sci 1989;46: 435᎐439. Chambers PA, Prepas EE, Gibson K. Temporal and spatial dynamics in riverbed chemistry: the influence of flow and
111
sediment composition. Can J Fisheries Aquatic Sci 1992;49:2128᎐2140. De Groot CJ, Golterman HL. Sequential fractionation of sediment phosphate. Hydrobiologia 1990;192:143᎐148. Demars BOL, Harper DM. The aquatic macrophytes of an English lowland river system: assessing response to nutrient enrichment. Hydrobiologia 1998;384:75᎐88. Denny P. Sites of nutrient absorption in aquatic macrophytes. J Ecol 1972;60:819᎐829. Denny P. Solute movement in submerged angiosperms. Biol Rev 1980;55:65᎐92. Garcıa-Ruiz R, Pattinson SN, Whitton BA. Denitrification in ´ river sediments: relationship between process rate and properties of water and sediment. Freshwater Biol 1998;39:467᎐476. Grasmuck L, Muller S. Assessment of the ¨ N, Haury J, Leglize ´ bio-indicator capacity of aquatic plants using multivariate analysis. Hydrobiologia 1995;300r301:115᎐122. Golterman HL, Sly PG, Thomas RL. Study of the relationship between water quality and sediment transport. Technical Papers in Hydrology 26 ŽA contribution to the International Hydrological Programme., UNESCO, Paris, 1983, 222 pp. Hesse PR. Phosphorus in lake sediments. In: Griffith EJ, Beeton A, Spencer JM, Mitchell DT, editors. Environmental phosphorus handbook. New York: John Wiley and Sons, 1973:573᎐583. Hieltjes AHM, Lijklema L. Fractionation of inorganic phosphates in calcareous sediments. J Environ Qual 1980; 9:405᎐407. Holmes NTH, Newman JR, Chadd S, Rouen KJ, Saint L, Dawson FH. Mean trophic rank: a user’s manual. R& D Technical Report E38. Bristol, UK: Environment Agency, 1999. Howard-Williams C. Cycling and retention of nitrogen and phosphorus in wetlands: a theoretical and applied perspective. Freshwater Biol 1985;15:391᎐431. Lijklema L. Dimensions and scales. Water Sci Technol 1998;37:1᎐7. Maitland PS. A simple corer for sampling sand and finer sediments in shallow water. Limnol Oceanogr 1969;14: 151᎐156. Martinova MV. Nitrogen and phosphor compounds in bottom sediments: mechanisms of accumulation, transformation and release. Hydrobiologia 1993;252:1᎐22. Murphy J, Riley JP. A modified single solution method for the determination of phosphate in natural waters. Anal Chim Acta 1962;27:31᎐36. Natural Environment Research Council ŽNERC.. Hydrological data United Kingdom: hydrometric register and statistics 1986᎐90. Wallingford: Institute of Hydrology, 1993. 179 pp.. Pella E, Columbo B. The study of carbon hydrogen and nitrogen by combustion-gas chromatography. Mikrochem Acta 1973;5:697᎐719. Preston CD, Croft JM. Aquatic plants in Britain and Ireland. Colchester, England: Harley Books, 1997:365.
112
S.J. Clarke, G. Wharton r The Science of the Total En¨ ironment 266 (2001) 103᎐112
Rose M. The role of sediment phosphorus concentration in controlling the growth of macrophytes in moving waters. Unpublished MSc Thesis, Dept. of Zoology, University of Leicester, 1995, 41 pp. Sand-Jensen K. Macrophytes as biological engineers in the ecology of Danish streams. In: Sand-Jensen K, Pedersen O, editors. Freshwater biology: priorities and development in Danish research, The Freshwater Biological Laboratory. Copenhagen: University of Copenhagen and G.E.C. Gad Publishers Ltd, 1997:74᎐101. Sand-Jensen K. Influence of submerged macrophytes on sediment composition and near-bed flow in lowland streams. Freshwater Biol 1998;39:663᎐679.
Sand-Jensen K, Jeppesen E, Nielsen K et al. Growth of macrophytes and ecosystem consequences in a lowland Danish stream. Freshwater Biol 1989;22:15᎐32. Schuurkes JA, Kok CJ, Den Hartog C. Ammonium and nitrate uptake by aquatic plants from poorly buffered and acidified waters. Aquatic Botany 1986;24:131᎐146. Stone M, English MC. Geochemical composition, phosphorus speciation and mass transport of fine-grained sediment in two Lake Erie tributaries. Hydrobiologia 1993;253:17᎐29. Svendsen LM, Rebsdorf A, Nornberg P. Comparison of methods for analysis of organic and inorganic phosphorus in river sediment. Water Res 1993;27:77᎐83.