ebook img

Riis T, Dodds WK, Kristensen PB, Baisner AJ. 2012. Nitrogen cycling and dynamics in a macrophyte PDF

13 Pages·2012·0.39 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Riis T, Dodds WK, Kristensen PB, Baisner AJ. 2012. Nitrogen cycling and dynamics in a macrophyte

FreshwaterBiology(2012)57,1579–1591 doi:10.1111/j.1365-2427.2012.02819.x Nitrogen cycling and dynamics in a macrophyte-rich stream as determined by a 15N-NHþ release 4 T. RIIS*, W. K. DODDS†, P. B. KRISTENSEN* AND A. J. BAISNER* *Department ofBioscience,Aarhus University, Aarhus, Denmark †DivisonofBiology, Kansas State University,Manhattan, KS,U.S.A. SUMMARY 1.A15N-NHþ tracerreleasestudywasconductedinamacrophyte-richstream,theRiverLilleaain 4 Denmark. The objectives of the study were to compare uptake rates per unit area of NHþ by 4 primary producers and consumers in macrophyte and non-macrophyte habitats, estimate whole- streamuptakeratesofNHþandcomparethistootherstreamtypes,andidentifythepathwaysand 4 estimatetherateatwhich NHþ entersthefoodwebinmacrophyteandnon-macrophytehabitats. 4 2. Macrophyte habitats had four times higher primary uptake rates and an equal uptake rate by primary consumers per unit habitat area as compared to non-macrophyte habitats. These rates represent the lower limit of potential macrophyte effects because the rates will be highly dependentonmacrophytebedheightandmeanbedheightintheRiverLilleaawaslowcompared to typical bed heights in many lowland streams. Epiphytes accounted for 30% of NHþ primary 4 uptake in macrophyte habitats, illustrating a strong indirect effect of macrophytes as habitat for epiphytes.Nfluxperunithabitatareafromprimaryuptakecompartmentstoprimaryconsumers was four times lower in macrophyte habitats compared to non-macrophyte habitats, reflecting much greater biomass accrual in macrophyte habitats. Thus, we did not find higher N flux from macrophyte habitats to primary consumers compared to non-macrophyte habitats. 3. Whole-stream NHþ uptake rate was 447 mgN m)2 day)1. On a habitat-weighted basis, fine 4 benthic organic matter (FBOM) accounted for 72% of the whole-stream uptake rate, and macrophytes and epiphytes accounted for 19 and 8%, respectively. 4.Wehadexpectedapriorirelativelyhighwhole-streamNuptakeinourstudystreamcompared to other stream types mainly due to generally high biomass and the macrophyte’s role as habitat for autotrophic and heterotrophic organisms, but our results did not confirm this. In comparison with other 15N-NHþ release study streams, we conclude that nutrient concentration is the overall 4 controllingfactorforNuptakeratesacrossstreams,mostlyasaresultofhighbiomassofprimary uptake compartments in streams with high nutrient concentrations in general and not in macrophyte streams in particular. 5. Our results indicate that macrophytes play an important role in the longer-term retention of N andthusadecreaseinnetdownstreamtransportduringthegrowingseasoncomparedtostreams without macrophytes, through direct and indirect effects on the stream reach. Direct effects are high uptake efficiency, low turnover rate (partly due to no direct feeding on macrophytes) and high longevity. An indirect effect is increased sedimentation of FBOM in macrophytes compared to non-macrophyte habitats and streams which possibly also increase denitrification. Increased retention with macrophyte presence would decrease downstream transport during the growing season and thus the N loading on downstream ecosystems. Keywords:lowland streams, macrophytes, nitrogencycling, nitrogen dynamics Correspondence:TennaRiis,DepartmentofBioscience,AarhusUniversity,8000AarhusC,Denmark.E-mail:[email protected] (cid:2)2012BlackwellPublishingLtd 1579 1580 T. Riis et al. invertebrates.Owingtothethree-dimensionalstructureof Introduction macrophyte beds, much of the biota in macrophyte Nitrogen (N) is a dominant pollutant in freshwater and streams is in intimate contact with the flowing water. marinesystems(Smith,2003).Agriculturalareasaremajor These features most likely influence stream N dynamics. contributors to N pollution (Dodds & Oakes, 2004) Additionally, there is some controversy about the extent because of the use of fertiliser and manure, and small to which rooted stream macrophytes use water column streams and rivers draining agricultural areas contribute nutrientsasopposedtosediment-derivednutrients(Mad- substantially to downstream N loading (Alexander et al., sen & Cedergreen, 2002). Herbivory on mature stream 2007). Lowland streams in agricultural catchments are macrophytes is mostly low (Elger, De Boer & Hanley, especially affected by N and are expected to contribute 2007), so it is not clear how intimately macrophytes are considerably to downstream transport of N pollutants. involved in N retention, N flux through lotic food webs Food-webdynamics,internalNcyclingandNretention andNuptakeinstreamswheretheydominate.However, have been studied in a variety of streams using 15N giventheirsubstantialbiomassandassociatedorganisms, releases,including atundra stream(Peterson et al.,1997), it is likely that macrophyte habitats do influence N foreststreams(Hall,Peterson&Meyer,1998;Mulholland dynamics to some degree. In part we are interested to et al., 2000; Tank et al., 2000; Ashkenas et al., 2004) and a knowwhethervascularmacrophytesserveasadeadend prairie stream (Dodds et al., 2000). Data from 12 such with respect to N uptake into the food web in a manner releasesindicatethatrelativelypristinesystemsarehighly similar to bryophytes in some streams (e.g. Dodds et al., retentive of N (Peterson et al., 2001; Webster et al., 2003). 2000). Less is known about N dynamics in N-rich systems (but The purpose of this study was to describe N cycling in see Royer, Tank & David, 2004). While much is known terms of whole-stream NHþ uptake, retention and flux 4 about short-term dynamics of N in small streams influ- into the food web of a lowland macrophyte-rich stream enced by agriculture (Mulholland et al., 2008), less is and to compare macrophyte and non-macrophyte habi- knownaboutfoodwebsonceNistakenup.Furthermore, tats. We added 15N over 12 days and quantified the 15N tracer studies in open lowland macrophyte-rich uptakeandturnover ratesof variousecosystemcompart- streamsarefew(butseeSimonet al., 2007),althoughthis ments for 80 days following the release. The overall type of stream is dominant in large parts of Europe and benefit of 15N release studies compared to 14N addition otherregionsaroundtheworld.Macrophyte-richstreams studies is the ability to trace the fluxes of N close to are common in low-gradient agricultural areas and are ambient concentrations (Peterson et al., 2001). Streams oftenaffectedbyanthropogenicsupplyofNtothestream with high N concentrations are particularly difficult to environment. Unique aspects of these streams, including studywithoutisotopicmethodsastheyaresaturatedwith high primary producer biomass and the influence of N, and perturbation of instream nutrient concentrations macrophytes on hydrodynamics raise the question leads to little observable effect on other ecosystem whether N cycling differs substantially from other types attributes (e.g. traditional measures of uptake length of streams. using nutrient additions do not work). Our specific Mosses and vascular macrophytes can be abundant in research questions were the following: What is the role somestreamsandlikelycontributetoNcycling.Lowland ofmacrophytesinNcyclingandlonger-termNretention? macrophyte-dominatedstreamsareoftencharacterisedby What is the role of macrophyte habitat versus non- a high biomass of aquatic vascular plants, including macrophytehabitat?Whatarewhole-stream uptake rates genera such as Potamogeton, Ranunculus and Myriophyl- of NHþ in macrophyte-rich streams and how does this 4 lum.Aquaticbryophytescanbeverybiologicallyactivein comparetootherstreamtypes?Atwhatrateandthrough some streams. In a deciduous forest stream (Walker which pathways does NHþ enter the food web in 4 Branch, Tennessee), ammonium uptake by the bryophyte macrophyteversusnon-macrophytehabitats?Wehypoth- Porellarepresented41%oftotalNretentionattheendofa esisedthatmacrophytehabitatsshowhigheruptakerates 6-week15Nadditionexperiment(Mulhollandet al.,2000). onanarealbasisthannon-macrophytehabitatsbecauseof Meyer&Likens(1979)recordedtheremovalofphospho- high macrophyte and associated epiphyte biomass. We rus when it passed over a bryophyte bed. The role of alsohypothesisedthatmacrophyte-richstreamswillshow vascularmacrophytesincyclingNinstreamsisevenless high whole-stream NHþ uptake rates because macro- 4 studied than the role of bryophytes. Macrophytes reduce phytes have a high active biomass and macrophyte water velocity (Sand-Jensen & Mebus, 1996) and create a streams have a generally high biomass of autotrophic large surface area that is colonised by epiphytes and compartments because of high nutrient concentration, (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 Nitrogen cycling in a macrophyte stream 1581 macrophytes provide a large biologically active surface significantdownstreamchangeinNHþconcentrationwas 4 area for epiphytic communities on leaves consisting of observed (data not shown), reinforcing the need for auto-andheterotrophicmicroorganismsandthestructure isotopic tracer methods in this stream. ofmacrophytescreatessubstantialsurfaceareaforwater- Stream water samples were analysed for total N on a biota contact. Finally, we hypothesised that N enters the Shimadzu TOC-Vcph including TNM-1 for N degrada- primary consumer food web at a higher rate in macro- tion. Samples were analysed by spectrometer for NHþ 4 phyte habitats than in non-macrophyte habitats, because by the phenol hypochlorite method (Greenberg, Clesceri macrophyte beds provide habitat for invertebrates and & Eaton, 1992). Analyses for NO(cid:2) were conducted 3 thus increase the biomass of primary consumers. using flow injection analysis with cadmium columns (LACHAT Instruments). Abundance of macrophytes in the study reach was determined by measuring length, Methods width and height of all macrophyte beds in the stream reach. Study site and experimental design We conducted the 15N release study in a 300-m reach in Biomass estimation of stream compartments the River Lilleaa situated in an agriculturally dominated Biomass of all compartments was determined at five catchment in eastern Jutland of Denmark (10(cid:3)03¢46.96¢¢E, stations located at equally spaced positions along the 56(cid:3)15¢00.83¢¢N). The catchment has mostly moraine soil experimental reach on 24 July (pre-sampling) and again and the run-off area is 72 km2, of which 74% is in onthe25 August(day30 after15N releasestarted).AllN agricultural production. The reach was primarily un- biomass values were subsequently habitat weighted by shaded with approximately a 20-m zone with overhang- surface area of inorganic substrate type (mud⁄clay, sand, ing trees. The remaining riparian areas were dominated gravel⁄cobble)ormacrophytecoverandvolume(seeData bycroplandandcattlegrazingwitha15-to30-msetback analysis). of weedy riparian vegetation in the area of study. The We measured the concentration of suspended particu- streamreachhadbeenchannelisedinthepastandismore lateorganicN(SPON)byfiltering0.5–1 Lofstreamwater or less straight without meanders. Our study design, at six locations each sample day during the experiment methods and data analyses generally follow Mulholland through pre-combusted Whatman GF⁄C filters, which et al. (2000) and Dodds et al. (2000). The study was were then dried and weighed. Samples were taken from designedto measure15Ntracer movementthroughbiotic downstream to upstream to avoid disturbance of sedi- and abiotic components in the stream during and after ments and contamination. 15N release. Prior to 15N addition, we measured stream We measured biomass of coarse benthic organic N physical and chemical properties, and biomass of major (CBON; >1 mm in size) and fine benthic organic N biologicalcompartments.15Nwasreleasedintothestream (FBON; <1 mm in size) at five locations in the reach. An for12 days.Before,duringandafter15Nrelease,samples open-ended PVC cylinder (16 cm in diameter; 201 cm2 from each major compartment were collected for 15N area) was pushed into the sediment. CBON was first analyses. Another measurement of biomass of major removed by hand and consisted typically of wood, ecosystemcompartmentswasperformedonemonthafter leaves and parts of emergent plants. The sediments were commencing the 15N release to account for changes of then lightly disturbed by hand down to about 1 cm biomass across the experimental period. depth, and the water in the cylinder was mixed. This sample represented surface FBON. A water sample of 200 mL was taken from the mixed cylinder water, Stream characteristics brought to the laboratory, filtered onto a pre-combusted Streamwidthwasmeasuredintransectsevery10 malong Whatman GF⁄C filter, dried and weighed. Deep FBON the 300-m experimental reach. In each transect, we was taken in a similar way but after raising the cylinder measured water depth and inorganic substrate type at slightly to allow the shallow FBON to be taken away by fivepointsacrossthestream.Aconservativesolute(NaCl) the current, then re-inserting the cylinder at the same release was performed to calculate mean travel time, location and creating a new hand disturbance down to mean water velocity, discharge and transient storage about 5 cm depth. parameters (Webster & Valett, 2006). A release of unla- Epilithon biomass was sampled at five locations along belled NHþ that doubled the stream NHþ concentration thestreamreachbyscrubbingandwashingallgraveland 4 4 wasperformedaweekbefore15Nrelease.Nostatistically stone material within the PVC core area. The collected (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 1582 T. Riis et al. material was filtered onto a pre-combusted Whatman feeding functional groups, and dry mass of each group GF⁄C filter, dried and weighed. was determined based on taxa biomass. Biomassofepiphytesandthetwodominantmacrophyte Fishbiomasswasestimatedbasedontwoelectrofishing species(RanunculusaquatilisLinn.andCallitrichesp.)were catchesaccordingtoSeber&LeCren(1967).Fishbiomass estimated based on macrophyte biomass taken using box was only measured once near the end of the experiment samples (inner dimensions: 17 cm · 11 cm · 12 cm). The (15August2010).Drymassvaluesweredeterminedafter box sampler was split horizontally along the centre at least 24 h in 60 (cid:3)C dry oven. allowing us to grab a sample in the macrophyte bed equivalent to the box sample volume. The volumetric Metabolism measures were later converted to area-specific mass as described in the data analysis section. We attempted to Whole-stream metabolism in terms of gross primary samplewithinmacrophytebedswithoutbiaswithrespect production (GPP) and community respiration (CR) was to depth or upstream⁄downstream location, but always determined 2 days before commencing the 15N release, sampled at least 5 cm from the stream bed to avoid N using the diurnal upstream-downstream dissolved oxy- associatedwithsediments.Aftercapturingasetvolumeof gen exchange technique (Marzolf, Mulholland & Stein- macrophyte, we used scissors to cut all the macrophyte man, 1994; Young & Huryn, 1998) modelled with the materials away around the perimeter of the box sampler. techniques described by Riley (2011). Concentrations of The sample included macrophytes, biofilm⁄sediment on dissolved oxygen were measured at 10-min intervals at the macrophytes and invertebrates living within the twolocationsonthereach(70and210 mfrom15Nrelease) macrophyte bed. Replicate box samples were taken from using YSI sondes (6600V2; Yellow Springs Instruments, representativeplacesintheupperpartofmacrophytebeds Yellow Springs, OH, USA). Reaeration rate was deter- distributedalongtheexperimentalreach.Oneboxsample mined using a propane method (Marzolf et al., 1994) was taken in each of five macrophyte beds of each of the coupled with salt (NaCl) release to ensure a plateau was twodominantspecies(totalof10samples). reached and to correct for any dilution. Photosynthetic Thesamplefromeachboxwasplacedinaplasticbagin active radiation (PAR) was measured using a Li-Cor the field and kept cool (4 (cid:3)C) until samples were quantum sensor and data logger (LI-COR 1400, LiCor processed (within 24 h following collection). In the labo- Biosciences, Lincoln, NE, USA). ratory, we separated macrophytes, epiphytes and inver- tebrates. First, epiphytes were washed off the 15N release, sampling and analysis macrophytes by gently squeezing in 2–3 batches of clean stream water, and all macrophyte material was removed A total of 50 g 98% enriched 15NH Cl (Batch no. 2181; 4 from the wash water. Then, the epiphyte biomass in the CamproScientific,Berlin,Germany)wereaddedoverthe wash water was filtered onto a pre-combusted Whatman 12-day release period commencing on 26 July 2010 and GF⁄C filter after removal of all macroinvertebrates, dried ending at 7 August 2010. A total of 13.761 g 15N were andweighed.Thecleanmacrophytematerialsweredried added to the stream. To accomplish this, a solution of in paper bags and weighed. 1.16 mMM 15NHþ4 was released into the stream at an Invertebrates from the box sampler represent the averagerateof6.9 mL min)1.Thisresultedinanaverage biomass of invertebrates associated with macrophyte concentration of 0.038 lg 15N L)1 in the stream water beds. Quantitative benthic macroinvertebrate samples with an average discharge 63 L s)1. Background NHþ 4 were collected with a small Surber sampler (22.1 cm2). concentration was 64.1 lg N L)1; thus, the addition of Threesampleswerepooledfromeachofthefivesites.For 15NHþ was <1% of total NHþ concentration. 4 4 each sample, sediments were disturbed to a depth of at Samplesofwater,organismsanddetrituswerecollected least 5 cm, and invertebrates were collected in a 250-lm from an upstream station, and six stations located 30, 73, net from the suspension made by the disturbance. 94,134,214and300 m downstreamfrom the 15N release. Samples were preserved in 96% ethanol until identifica- Sampling of all biological compartments was carried out tion. thedaybeforerelease,duringrelease(days1,3,6,11)and In the laboratory, invertebrate specimens were identi- afterthereleaseperiod(days13,15,18,23,46,92). fied to the lowest possible taxon. Number of individuals Qualitative samples for 15N analyses of epilithon, wascountedforeachtaxonineachsample,drymasswas epiphytes, detritus, fine benthic organic matter (FBOM) measured and a mean dry mass per individual was and macrophytes were taken in the same way as estimated. Invertebrate taxa were then separated into described for biomass but in small amounts, working a (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 Nitrogen cycling in a macrophyte stream 1583 few metres upstream on each subsequent sampling date Metabolismwascalculatedwithamodelthataccounted toavoidre-samplingdisturbedareas.Samplesoffishand for light, travel time, aeration and temperature effects. each major functional feeding group of macroinverte- This model minimised sums of squares of observed and brates were taken at each station by netting and hand calculated dissolved oxygen concentrations through an picking.Allsolidsampleswerereturnedtothelaboratory iterative fitting function (Riley, 2011). Transient storage on ice, dried at 60 (cid:3)C and ground before analysis. was calculated using OTIS solute transport model soft- Invertebrate and fish samples were freeze-dried. Macro- ware (Runkel, 1998). invertebratesampleswereleftat4 (cid:3)Cinwaterfor12 hso Calculations of d15N in samples, uptake lengths and they would empty their guts before they were dried and rates, nitrification, regeneration and retention followed ground for analysis. the procedures described by Mulholland et al. (2000), Thewaterwassampledandanalysedfor15N-NHþ and Dodds et al. (2000) and Ashkenas et al. (2004). Inter- 4 15N-NO(cid:2) 1 dayafterthereleasebegantoestimateuptake conversion among areal uptake rates, turnover rates, 3 of NHþ with minimal interference from mineralisation turn-over lengths and uptake velocity were calculated 4 and to measure nitrification rates. The 15NHþ was also accordingtotheequationspublishedbytheStreamSolute 4 sampledonedayafterthereleasewasstoppedtocalculate Workshop (1990). Whole-stream spiral length was calcu- therateofregenerationof15Nbacktothewater.15N-NHþ lated from exponential decline in 15N-NHþ downstream 4 4 inthewaterwasisolatedforanalysisusinganammonium from the release point on day 1. Uptake rates of NHþ-N 4 diffusion method suggested by Sørensen & Jensen (1991) by each primary uptake compartments were calculated and Holmes et al. (1998) and described in detail in from 15N values on day 3 at station 2–4 and the tracer Mulholland et al. (2000). 15N-NO(cid:2) was isolated and 15N⁄14N ratios in stream water NHþ at those stations on 3 4 analysed as described previously (Mulholland et al., day 0, following Mulholland et al. (2000) and Tank et al. 2000; Peterson et al., 2001). (2000).Thevalueswerecorrectedforturnoverlossof15N Solid samples were dried and finely ground before from each compartment during the first three days analysis for 15N⁄14N ratios by the Stable Isotope Mass (Mulholland et al., 2000). We also calculated uptake rates Spectrometry Laboratory at Kansas State University on a per unit habitat area in macrophyte habitats and non- Thermo Finnigan Delta Plus mass spectrometer. The macrophyte habitats adjusting to the different compart- standard deviation range for 15N was 0.03–0.18& based ments present in macrophyte and non-macrophyte on standards every 12th sample. Analysis of un-enriched habitats, respectively. Turnover rates of N in primary replicatesamplesfromthisstudyindicatedlowanalytical uptake compartments were calculated as the exponential variation(<3%coefficientofvariation).The15N-NHþwas decline of 15N (depuration) from 1 to up to 49 days after 4 analysed following an alkaline diffusion procedure 15N drip stop. Nitrogen retention was calculated as mass (Holmes et al., 1998) on the same mass spectrometer as balance of the total 15N added and retained over the described above. Additional samples were processed for experimental period. Retention of each compartment analysis of 15N-NO(cid:2) by boiling under basic conditions to withinthe300 mreachwasbasedondownstreamdecline 3 concentrate NO(cid:2) and drive off all NHþ as NH . Then, incomponent-specific15Nbiomassonday11,thelastday 3 4 3 15N-NO(cid:2) + NO(cid:2) was converted to ammonia with De of isotope release. If the slope of the regression of 15N 3 2 Varda’sAlloy,concentratedbydiffusion,andanalysedas biomass as a function of distance was not significant above (Sigman et al., 1997). (P < 0.05), we used the mean 15N biomass for the reach. Uptakeandturnoverratesofconsumerswereestimated usingaboxmodelthatcanaccountforchanginglabelina Data analysis food source over time (Dodds et al., 2000). This approach Biomass of benthic compartments was calculated as concentratesonasinglestationovertimeandusestrends habitat-weighted values. Mass in the reach of each in measured label in food sources, and biomass of the macrophyte species was determined using percentage uptake compartment, to calculate a rate of label uptake coverandmassperunitareavalues.Massofinvertebrates intothe consumer compartmentbasedon anuptakerate. associated with macrophytes was based on mass in Food sources for each functional feeding group were volume of macrophytes because of the importance of the determined using Wallace & Webster (1996) and Skirver three-dimensionalstructureofthishabitat.Volume-based (1982), and proportion of each food source was based on masses were then multiplied by mean height of macro- priorexperience.Theuptakerateisvariedtominimisethe phytebedsinthereach(0.13 m)toconverttomassperm2 sum of square of error between observed and calculated stream reach. consumer isotopic contents. In some cases, the consumer (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 1584 T. Riis et al. compartment was more labelled than the putative food Table1 Ecosystempropertiesintheexperimentalstreamreachin sources. In this case, the pattern of labelling in the food theRiverLilleaaduringtheexperiment.Datagivenasmean±SD wherepossible compartmentwaspreservedbutamultiplierwasusedto optimise the fit. The multiplier accounts for the selective Ecosystemproperty(unit) Value nature of animal feeding. This additional multiplier was Reachlength(m) 300 fitted simultaneously with the uptake rate in these cases. Averagewidth(m) 2.6±0.2 The fitting was performed using the Solver option in Averagedepth(m) 0.23±0.08 Microsoft Excel. Values given are based on one best or Totalsurfacearea(m2) 709 Gradient(mkm)1) 1.2 mean of two best stations. %run 94.3 We included a meta-analysis to compare our macro- %riffle 5.7 phyte stream to other stream types. We collated existing %mudandclay 18.8 data on nutrient dynamic parameters from 15N release %sand 46.9 %gravel⁄cobble 34.4 studies(Websteret al.,2003;Simonet al.,2007)andrelated %totalmacrophytecover 10.7 them by linear regression to NHþ concentration in the 4 Meanmacrophyteheight(cm) 14 streams to compare NHþ4 uptake parameters in our %totalmacrophytevolume 8.0 macrophyte-richstreamwithotherstreamtypes.Wealso %coverRanunculusaquatilis 8.2 determined the linear relationship between NHþ in %coverCallitrichesp. 2.0 4 %coverBerulaerecta 0.5 stream water and biomass of primary uptake compart- Meandischarge(Ls)1) 63.2 ments, and the linear relationship between biomass of Meanwatervelocity(ms)1) 0.10 primary uptake compartments and uptake rate. Strong Meantraveltime(min) 46 relationships between NHþ uptake parameters and NHþ TransientstorageAs⁄A 0.02 4 4 Trans.stor.exc.coef.(a;K) 4.50·10)5 concentration were previously found within streams 1 Meanwatertemperature((cid:3)C) 12.4±1.2 (Wollheim et al., 2001; Dodds et al., 2002) and among a NHþ (lgNL)1) 64.1±6.2 4 range of streams (Webster et al., 2003). NO(cid:2) (lgNL)1) 1433.3±70.3 3 TotalN(mgNL)1) 2.1±0.04 PO3(cid:2) (lgPL)1) 63.3±9.7 4 TotalP(lgPL)1) 88.5±18.0 Results TOC(mgCL)1) 3.0±0.1 pH 7.90 Environmental properties Alkalinity(mEqL)1) 2.87±0.34 Grossprimaryproduction(gO m2day)1) 1.65 Physical characteristics in the stream were typical for 2 Respiration(dielO changemethod,gO m2day)1) 5.29 2 2 summer conditions in small low-gradient macrophyte- P⁄Rratio 0.21 rich streams (Table 1). We did not detect any significant groundwater input along the reach, based on NaCl releases, thus simplifying most calculations. Bottom sub- strata areas were dominated by sand and gravel. Macro- Table2 Habitat-weightedNbiomassofecosystemcompartments phyte cover was 11% in the experimental reach and orderedbytrophiclevel.Averagevaluesbasedontwomeasure- dominated by R. aquatilis L. and Callitriche sp. (Table 1). mentsat24Julyand23August,respectively Meantravel time of waterthroughthe reach was46 min, Biomass C:N and relative transient storage area (storage area as Compartment (mgNm)2) %N (molar) proportion of profile area; A ⁄A, m2 m)2) was 0.02. Both s N and phosphorus content in the water were relatively SPON 48 3.3 11.6 Epiphytes 634 11.5 10.6 high (Table 1). Gross primary production was one-third Epilithon 208 6.2 8.8 the diurnal respiration, indicating that the stream is net Macrophytes 1475 3.7 11.9 heterotrophic despite the high macrophyte cover. FBONsurface 22701 6.0 13.0 FBONdeep 23872 4.8 13.6 Detritus 375 1.2 38.3 Nitrogen compartments Grazers 39 8.5 5.4 Collectors 89 9.9 5.0 Nitrogen stocks were dominated by FBON that consti- Gammaruspulex 115 7.4 6.0 tuted 94% of the total N biomass (Table 2). Macrophyte Shredders 72 7.4 7.9 Predators 32 9.6 6.2 biomass supported high epiphyte biomass, which three Fish 2808 12.6 4.3 times higher than epilithon biomass. Gammarus pulex (L.) (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 Nitrogen cycling in a macrophyte stream 1585 was the dominant invertebrate on the reach and consti- (Fig. 1). Epiphytes were responsible for one-third of the tuted36% of the primary consumer N biomass(Table 2). uptakeinthemacrophytehabitat.Inthenon-macrophyte Biomass of invertebrates in macrophyte habitats was habitat, FBOM was responsible for 98% NHþ uptake. 4 about one-third higher than non-macrophyte habitats Whole-stream spiral length was 303 m, resulting in a (Table 3). Macrophyte habitats were especially important uptake velocity (v) of 0.08 mm s)1. Whole-stream uptake f for Simulidae, shredders (dominated by Limnephilidae) rate expressed as uptake rate of water 15NHþ was 4 and G. pulex. 447 mg N m)2 day)1. The measured compartment-spe- cific NHþ uptake rates based on incorporation of 15NHþ 4 4 from the water column indicated FBON was responsible Nitrogen uptake rates for 72% of the daily uptake rates (Table 4). Macrophytes Macrophyte habitats in the River Lilleaa showed fivefold and epiphytes were only responsible for 19% and 8% of higher NHþ uptake rates than non-macrophyte habitats whole-stream NHþ uptake, respectively (Table 4). Mac- 4 4 rophyteswerethemosthighlylabelledatday11ofrelease Table3 ComparisonofNbiomass(gNm)2)inmacrophytehabitats of all primary uptake compartments and about 2.5 times (mixedRanunculusaquatilisandCallitrichesp.)andnon-macrophyte higher than epilithon (Fig. 2a,b). When we summed the habitatsintheRiverLilleaa.ValuesareaveragesofsamplesfromJuly andAugust2010.Thevaluesshowtheimportanceofmacrophytesas compartment-specificNHþ uptakerate,wecouldaccount 4 habitatsforinvertebratesandespeciallySimulidae,Gammaruspulex for50%ofwholereachNHþuptake(measuredbydecline 4 andshredders of 15NHþ in the water column). Biomass-specific uptake 4 Macrophyte Non-macrophyte ratesshowtheefficiencyofthedifferentcompartmentsto habitatLilleaa habitats takeupNHþ.Macrophytesandtheirassociatedepiphytes FFG (mgNm)2) (mgNm)2) 4 have threefold higher biomass-specific uptake rates than Grazers 5.1 42.4 epilithonand10timeshigherthanFBON(Table 4).There Collectorscomp 15.2 96.7 was no trend in 15N-NO(cid:2) in water down through the 3 Simulidae 110.1 55.1 experimental reach at day 12 and thus nitrification was Shredders 148.1 62.8 not detectable, although rates must be less than the total G.pulex 247.5 99.2 Predators 6.6 34.3 rate of 15N-NHþ4 uptake. Total 532.7 390.4 Table4 Habitat-weighteduptakeratesofNHþ,biomass-specific 4 uptakerates,turnoverratesand%15Nretainedoftotalretentionfor Shredders Collectors Shredders Collectors majorecosystemcompartmentsinLilleaa.UptakeratesforNO(cid:2) are 3 5.3 19.7 2.3 24.0 estimatedasthedifferencebetweentotalNuptakeandNHþuptake 4 Grazers Gammarus pulex Grazers Gammarus pulex (seetextforfurtherexplanation).Biomass-specificuptakerateis 0.7 10.0 5.7 4.0 calculatedasuptakerate(mgNm)2day)1)⁄compartmentbiomass (mgNm)2) 35.7 36.0 Biomass- Surface FBON Deep FBON NHþ specific NO(cid:2) 4 3 Macrophytes Epiphytes 86.5 74.3 uptake NHþ uptake %15N 4 444.0 191.6 Epilithon CBON rate uptake Turnover rate retained 3.0 0.6 Ecosystem (mgN rate rate (mgN oftotal compartment m)2day)1) (day)1) (day)1) m)2day)1) retention Macrophyte habitat Non-macrophyte habitat Epiphytes 18 0.032 0.042 12 7.6 635.5 164.4 Epilithon 2.3 0.011 0.081 14 0.2 Macrophytes 42 0.027 0.027 0 3.0 NH4+ FBONsurface 86 0.004 0.017 – 50.8 FBONdeep 74 0.003 0.102 – 31.9 Detritus 0.6 0.002 0.015 5 0.7 Fig.1 Diagramofuptakeratesperunithabitatarea Grazers 5.3 0.136 0.135 – 0.1 (mgNm)2day)1)inmacrophyteandnon-macrophytehabitatsin Collectors 14 0.157 0.158 – 0.7 theRiverLilleaa.UptakeratesofNHþ-Nbyeachprimaryuptake 4 Gammarus 4.7 0.041 0.041 – 0.2 compartmentswerecalculatedfrom15Nvaluesonday3atstation pulex 2–4andthetracer15N⁄14NratiosinstreamwaterNHþ atthose 4 Shredders 2.6 0.036 0.036 – 0.1 stationsonday0,followingMulhollandetal.(2000)andTanketal. Predators 0.5 0.016 0.014 – 0.03 (2000).Uptakeratesforprimaryconsumersarefromaboxmodel Fish 22 0.008 0.008 – 4.8 describedinRiley(2011). (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 1586 T. Riis et al. (a) 250 Table5 Statisticalparametersforthreemodelstestedontherela- R. aquatilis tionshipbetweenarealuptakerate(U)andNHþconcentration(data Epi Ran 4 showninFig.4).Modelsarethelinearmodel,theefficiencyloss 200 Callitriche modelandtheMichaelis–Mentensaturationmodelasdescribedin Epi Cal O’Brienetal.(2007) ‰) 150 Model Equation R2 P N ( δ 100 Linear:y=ax+b U=0.007x+0.033 0.79 0.000 Efficiencyloss:y=axb U=0.002*x0.774 0.77 0.049 Michaelis–Menten: U=1.389x⁄(136.9+x) 0.75 0.054 50 y=ax⁄(b+x) Turnoverratesof15Nwerewithinthesamerangeforall 0 0 10 20 30 40 50 60 70 autotrophiccompartmentsalthoughmacrophyteturnover Time (days) time was three timesslower than epilithon (Table 4). The level of 15N enrichment in all primary uptake compart- (b) 100 SPON ments except macrophytes declined quickly after Epilithon 15N-NHþ addition was stopped (Fig. 2a,b). Enrichment 4 80 sFBON of15Ninmacrophytespeakedaftertheadditionstopped, dFBON Detritus stayed high in the plants for longer than most other ‰) 60 primary uptake compartments and was almost back to N ( background 15N values 48 days after the isotope release δ 40 ended (Fig. 2a). The same occurred for detritus (Fig. 2b). The 15N retention in primary uptake compartments was 20 alsoreflectedintheprimaryconsumersthatwerestill15N enriched after 96 days after drip stop (Fig. 2c). Turnover rate of NHþ in the water was 31 day)1, and SPON and 0 4 0 10 20 30 40 collectors had the highest turnover rate of any compart- Time (days) ment other than the NHþ, at 0.19 day)1 and 0.16 day)1, 4 respectively. Gammarus pulex, shredders and detritus had (c) 160 Gammarus pulex lowturnoverrates,andpredatorsandfishhadthelowest. 140 Collectors composite If we assume that biomass does not change, then total 120 Shredders Grazers loss rate of N (mineralisation) should equal total uptake. 100 Total uptake, assuming organic N uptake is minimal, is ‰) 80 thesumofNO(cid:2) andNHþ uptake.Thus,wecancalculate N ( compartment-s3pecific tot4al N uptake, NHþ uptake, and 15 60 4 δ the difference should represent the compartment-specific 40 NO(cid:2) uptake. We are only comfortable using this calcu- 3 20 lation for compartments that are in intimate contact with 0 the water column because the NHþ uptake calculation 4 depends on knowing the 15NHþ content of the water. 4 40 Estimations suggest that NO(cid:2) uptake was considerable Predators inv. 3 ‰) 30 Fish andinmostcasesasimportantasormoreimportantthan N ( 20 NHþ4 uptake (Table 4). The estimations showed that no δ15 10 NO(cid:2)3 was taken up by macrophytes. Attheendofthe15Nreleaseatday11,atotalof26%of 0 0 10 20 30 40 50 60 70 80 90 100 110 theaddedtracerwasretainedinthestreamreach.Mostof Time (day) the retained 15N was retained in FBON (83%), and macrophytes and epiphytes were responsible for 8% of Fig.2 Timeseriesofobservedd15Natstation3(94mfromrelease the retained 15N (Table 4). The significant but not com- site)for(a)macrophytesandepiphytes,(b)otherprimaryuptake plete retention of added 15N is consistent with the 303 m compartments,and(c)primaryandsecondaryconsumers.Thebro- kenlinemarksthetime(day12)when15NHþadditionwasstopped. calculated uptake length for NHþ. 4 4 (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 Nitrogen cycling in a macrophyte stream 1587 Uptake rates of 15N by primary consumers in Lilleaa predators feed primarily on Simulidae. Grazers had were similar in macrophyte and non-macrophyte habi- lower natural 15N than their presumed food source tats (Fig. 1). In macrophyte habitats and non-macro- (epilithon), most likely reflecting a missing food source phyte habitats, respectively, 6 and 22% of the primary in the sampling or physiological discrimination of food uptake was taken up by primary consumers. On a source by the grazers. whole-stream basis, total N flux into primary consumers constituted 16% of the primary producer’s uptake in Meta-analysis of whole-stream uptake parameters Lilleaa (total uptake rate of primary consumers as proportion of total uptake rate by primary uptake We compared whole-stream uptake parameters for NHþ 4 compartments; Table 4). in our macrophyte stream to other stream types (Fig. 4). The parameters in the River Lilleaa were measured at twice the NHþ concentrations found in other streams we Food-web relations 4 usedforcomparison.UptakelengthinLilleaawasshorter To determine whether macrophytes are directly food than some streams with lower NHþ concentration 4 source for invertebrates in the River Lilleaa, we analy- (Fig. 4a),butuptakevelocitywassimilartootherstreams sed the natural abundance of 15N in primary and with lower NHþ concentrations (Fig. 4b). Areal uptake 4 secondary uptake compartments (Fig. 3). Epiphytes, rates in Lilleaa and two New Zealand macrophyte epilithon, SPOM and detritus showed similar 15N values streams were at the same level as other stream types around 6&, and FBON was 4–5&. Macrophytes had (Fig. 4c). greater 15N values than the other primary uptake compartments and greater than the primary consumers, the latter reflecting that there was little direct feeding on the macrophytes. Among the primary consumers, the (a) Sw = 12.672*x1.0828 filterer Simulidae had the lowest natural 15N value. R2 = 0.598; P < 0.05 1000 Gammarus pulex and predators were close and lower m) than Baetidae, Ephemera and shredders. This probably (w S indicates that the omnivore G. pulex to a high degree 100 feeds on FBON and primary producers, and invertebrate (b) 10 1) – s m 0.1 m v (f (c) 0.01 U = 0.017*x0.774 1) R2 = 0.779; P < 0.0001 – y a d –2 0.1 m N g U ( 0.01 1 10 100 NH + concentration (µg L–1) 4 Fig.4 Relationshipsofuptakelength(S ;a),uptakevelocity(m;b) w f anduptakerate(U;c)asafunctionofwaterNHþconcentrationina 4 rangeofstreamtypes.DatafromLilleaa(a,bandc)andthetwo Fig.3 Naturalabundanceformaincompartmentsarrangedafter macrophytestreamsreportedinSimonetal.(2007;c)arecircled. presumedtrophiclevel.Valuesareaveragefrombackgroundsam- DatafromotherstreamscollatedfromWebsteretal.(2003)and ples. Simonetal.(2007).Notelog-logtransformation. (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591 1588 T. Riis et al. Discussion uptakeinthehighlyagriculturallyimpactedRiverLilleaa. In our study, abiotic and biotic NHþ uptake by surface Macrophyte habitat versus non-macrophyte habitats 4 FBOM was released back into the water after about eight Macrophyte habitat showed four times higher NHþ weeks,correspondingtofindingsinPetersonet al.(2001). 4 uptake rates in the River Lilleaa than non-macrophyte DeepFBOM(3–5 cmsedimentdepth)hadafivefoldfaster habitats. This represents a lower level for uptake rate in turnover rate than surface FBOM, possibly due to deni- macrophytehabitats becauseofthe relative lowheight of trification in this depth zone (Stelzer et al., 2011). macrophyte beds in the study river. In deeper streams, Our second hypothesis that N enters the primary macrophyte beds are often more than 0.5 m (Riis, Sand- consumer food web at a higher rate in macrophyte Jensen & Larsen, 2001). Provided that uptake is evenly habitats than in non-macrophyte habitats, because mac- distributed throughout the vertical dimension of macro- rophyte beds provide habitats for invertebrates and thus phyte beds, the uptake rates can therefore easily exceed increase the density of primary consumers, was partly theuptakeratefoundinthisstudy.Asanexample,ifthe verified. The 15N uptake rate by primary consumers was meanheightofmacrophytebedsintheRiverLilleaawere equal in the macrophyte and non-macrophyte habitats in 0.5 minsteadof0.14 m,theprimaryuptakeratewouldbe the River Lilleaa, because invertebrate biomass is almost about four times higher than the actual value in macro- equal in the two habitats. This was surprising because phyte habitats and consequently 14 times higher than in invertebrate density generally increases with increasing non-macrophyte habitats. Direct uptake by macrophytes habitat area of macrophytes (e.g. Strayer & Malcolm, was responsible for 70% and epiphytes for 30% of the 2007). However, the low macrophyte bed height in the total uptake in macrophyte habitats, showing a relatively River Lilleaa limits the difference in invertebrate density large indirect effect of macrophytes through their role as between macrophyte and non-macrophyte habitats. For habitats for epiphytes. Ammonium uptake efficiency example,ifmacrophytemeanheighthadbeen0.5 mthere expressed as biomass-specific uptake rate of NHþ was would have been about four times higher invertebrate 4 higher in both macrophytes and epiphytes compared to biomass and consequently four times higher uptake rate the uptake compartments in non-macrophyte habitats. by primary consumers in macrophyte habitats compared Biomass-specific uptake rate for macrophytes in the to non-macrophyte habitats, given equally distributed River Lilleaa (0.027 day)1) corresponds to biomass-spe- uptakebyprimaryconsumersthroughoutthemacrophyte cific uptake rates measured in macrophytes in two New bed. However, detailed studies on N uptake within Zealand (NZ) streams (0.021 and 0.032 day)1), although macrophyte beds are needed to validate a linear up- the dominant species was Nasturtium officinale in the NZ scaling of invertebrate uptake in macrophyte beds of streams and R. aquatilis and Callitriche sp. in Lilleaa. This varying sizes. indicates a relatively stable uptake efficiency among The large discrepancy between uptake by primary macrophyte species and even across growth forms uptake compartments and flux into primary consumers because N. officinale is an amphibious species whereas in macrophyte habitats most likely reflects high biomass R. aquatilis and Callitriche sp. are submerged species (Riis accrual by macrophytes and lack of feeding in macro- et al., 2001). We did not measure uptake in the sediment phytesbyconsumers.Inmacrophytehabitats,only6%of undermacrophytebeds,butifweassumeitwasthesame the NHþ taken up was passed on to primary consumers, 4 as in open habitats, then total N retention in macrophyte comparedtonon-macrophytehabitatswhere22%moved habitats will be about 10% higher than estimated here into primary consumers. Although this could indicate because of the c. 10% cover of macrophytes in the reach. high mineralisation directly from the primary producer’s Uptake rates of NHþ in non-macrophyte habitats were uptake, the turnover rates of macrophytes are similar to 4 highly dominated by FBOM, which is a mix of direct othercompartmentsandasubstantialpartofthediscrep- uptake by autotrophic and heterotrophic microalgae and ancy is most likely due to build-up of biomass in sorption onto sediments. Richey, McDowell & Likens macrophytes. Our study confirms that more 15N persist (1985) concluded that sorption of NHþ to sediments was longer in macrophytes than other primary uptake com- 4 responsible for significant storage during summer and partments, reflecting substantial biomass accrual. autumn. We were not able to separate abiotic and biotic Natural 15N abundances in macrophytes were much uptake in FBOMin this study but this is unimportant for more enriched than microalgae in our study. The isotope estimating total N retention. Uptake rates by FBOM natural abundance could reflect that macrophytes use generally increase with increasing agricultural develop- sediment nutrients that are affected by enriched ground ment (Simon et al., 2007), supporting the high FBOM water in addition to stream water nutrients. Thus, (cid:2)2012BlackwellPublishingLtd,FreshwaterBiology,57,1579–1591

Description:
Nitrogen cycling and dynamics in a macrophyte-rich stream as determined primary consumers per unit habitat area as compared to non-macrophyte habitats.
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.