J. N. Am. Benthol. Soc., 1988, 7(4):410-432 ffl 1988 by ...coweeta.uga.edu/publications/664.pdfmass...
J. N. Am. Benthol. Soc., 1988, 7(4):410-432 ffl 1988 by The North American Benthological Society Elemental dynamics in streams* JUDY L. MEYER', WILLIAM H. McDowELL 2 , THOMAS L. Borr 3 , JERRY W. ELWooo 4 , CHANEL ISHIZAKI S , JOHN M. MELACK S , BARBARA L. PECKARSKY 7 , BRUCE J. PETERSON", AND PARKE A. RUBLEE' 1 Institute of Ecology and Zoology Department, University of Georgia, Athens, Georgia 30602 USA 2 Research Center, King Hall, SUNY Oswego, Oswego, New York 13126 USA 3 Stroud Water Research Center, Academy of Natural Sciences, R.D. 1, Box 512, Avondale, Pennsylvania 19311 USA 4 Environmental Sciences Division, Oak Ridge National Lab, Oak Ridge, Tennessee 37831-6036 USA 5 Institute Venezolano de Investigaciones Cientiftcas, Apartado 21827, Caracas 1020-A, Venezuela 6 Department of Biological Sciences, University of California, Santa Barbara, California 93106 USA 7 Department of Entomology, Cornell University, Ithaca, New York 14853 USA 8 Ecosystems Center, Marine Biological Laboratory, Woods Hole, Massachusetts 02540 USA 9 Biology Department, Whitman College, Walla Walla, Washington 99362 USA Abstract. We discuss elemental dynamics in streams and seek to identify areas where there are critical gaps in our understanding. Both landscape-level processes (e.g., geology, land-use practices, vegetation) and heterogeneous in-stream processes influence the supply and availability of elements to the stream biota. Stream ecologists need to consider the relative availability of different com- pounds or groups of compounds to the biota rather than lumping all forms of an element into operationally-defined units such as dissolved organic nitrogen or carbon. The impact of short-term events like storms on the elemental dynamics in streams needs to be assessed and compared with other controls. The relative importance of longitudinal (upstream), lateral (riparian zone, flood- plains), and in-stream controls of supply and availability of elements needs to be compared in a variety of streams. Availability of essential elements is a key factor controlling rates of primary productivity and decomposition in streams. ^hole system manipulations offer a valuable tool for understanding the interactions between elements and all components of the stream food web. We include an action plan of developments that would assist researchers in addressing some of the critical gaps we have identified in our understanding of elemental dynamics in streams. Key words: nutrients, elements, streams, spiraling, dissolved organic carbon, nitrogen, phos- phorus. In this paper we discuss ideas on elemental tions for new programs that could stimulate dynamics in streams and point out critical gaps progress in stream ecology. This paper is not in our understanding of the dynamics of dis- intended as a comprehensive review of ele- solved organic carbon (DOC) and dissolved and mental or organic matter dynamics in streams; particulate forms of nutrients such as nitrogen our intent is to stimulate further discussion and and phosphorus in both streams and rivers. We new research. first discuss controls on the supply and avail- There are several compelling reasons for con- ability of elements in streams and then examine sidering cycling of elements in streams. First, processes that are regulated by supply and to the extent that nutrients are limiting in availability. We conclude with some sugges- streams, they regulate the rates at which im- portant ecological processes such as primary * Paper resulting from a Working Group discussion Productivity or decomposition proceed; changes at a symposium on "Community structure and func- in these ratesresult in alterations of stream com- tion in temperate and tropical streams" held 24-28 munity structure (e.g., Elwood et al. 1981, Pe- April 1987 at Flathead Lake Biological Station, Uni- terson et al. 1985). Changes in macroinverte- versity of Montana, Poison, USA. brate community structure will also have an 410
J. N. Am. Benthol. Soc., 1988, 7(4):410-432 ffl 1988 by ...coweeta.uga.edu/publications/664.pdfmass by grazers will influence rates of nutrient uptake. The interaction between elemental
Text of J. N. Am. Benthol. Soc., 1988, 7(4):410-432 ffl 1988 by ...coweeta.uga.edu/publications/664.pdfmass...
J. N. Am. Benthol. Soc., 1988, 7(4):410-432ffl 1988 by The North American Benthological Society
Elemental dynamics in streams*
JUDY L. MEYER', WILLIAM H. McDowELL2, THOMAS L. Borr3,JERRY W. ELWooo4, CHANEL ISHIZAKIS, JOHN M. MELACKS,
BARBARA L. PECKARSKY7, BRUCE J. PETERSON", AND PARKE A. RUBLEE'1 Institute of Ecology and Zoology Department, University of Georgia,
Athens, Georgia 30602 USA2 Research Center, King Hall, SUNY Oswego, Oswego, New York 13126 USA
3 Stroud Water Research Center, Academy of Natural Sciences, R.D. 1, Box 512,Avondale, Pennsylvania 19311 USA
4 Environmental Sciences Division, Oak Ridge National Lab,Oak Ridge, Tennessee 37831-6036 USA
5 Institute Venezolano de Investigaciones Cientiftcas, Apartado 21827,Caracas 1020-A, Venezuela
6 Department of Biological Sciences, University of California,Santa Barbara, California 93106 USA
7 Department of Entomology, Cornell University, Ithaca, New York 14853 USA8 Ecosystems Center, Marine Biological Laboratory, Woods Hole, Massachusetts 02540 USA
9 Biology Department, Whitman College, Walla Walla, Washington 99362 USA
Abstract. We discuss elemental dynamics in streams and seek to identify areas where there arecritical gaps in our understanding. Both landscape-level processes (e.g., geology, land-use practices,vegetation) and heterogeneous in-stream processes influence the supply and availability of elementsto the stream biota. Stream ecologists need to consider the relative availability of different com-pounds or groups of compounds to the biota rather than lumping all forms of an element intooperationally-defined units such as dissolved organic nitrogen or carbon. The impact of short-termevents like storms on the elemental dynamics in streams needs to be assessed and compared withother controls. The relative importance of longitudinal (upstream), lateral (riparian zone, flood-plains), and in-stream controls of supply and availability of elements needs to be compared in avariety of streams. Availability of essential elements is a key factor controlling rates of primaryproductivity and decomposition in streams. ^hole system manipulations offer a valuable tool forunderstanding the interactions between elements and all components of the stream food web. Weinclude an action plan of developments that would assist researchers in addressing some of thecritical gaps we have identified in our understanding of elemental dynamics in streams.
In this paper we discuss ideas on elemental tions for new programs that could stimulatedynamics in streams and point out critical gaps progress in stream ecology. This paper is notin our understanding of the dynamics of dis- intended as a comprehensive review of ele-solved organic carbon (DOC) and dissolved and mental or organic matter dynamics in streams;particulate forms of nutrients such as nitrogen our intent is to stimulate further discussion andand phosphorus in both streams and rivers. We new research.first discuss controls on the supply and avail- There are several compelling reasons for con-ability of elements in streams and then examine sidering cycling of elements in streams. First,processes that are regulated by supply and to the extent that nutrients are limiting inavailability. We conclude with some sugges- streams, they regulate the rates at which im-
portant ecological processes such as primary* Paper resulting from a Working Group discussion Productivity or decomposition proceed; changes
at a symposium on "Community structure and func- in these rates result in alterations of stream com-tion in temperate and tropical streams" held 24-28 munity structure (e.g., Elwood et al. 1981, Pe-April 1987 at Flathead Lake Biological Station, Uni- terson et al. 1985). Changes in macroinverte-versity of Montana, Poison, USA. brate community structure will also have an
1988] ELEMENTAL DYNAMICS IN STREAMS 411
effect on elemental dynamics; e.g., the main-tenance of high algal productivity but low bio-mass by grazers will influence rates of nutrientuptake. The interaction between elemental andcommunity dynamics is a major focus of thisdiscussion.
A second reason to consider elemental dy-namics in streams is that elements link aquaticand terrestrial ecosystems, and in-stream pro-cesses are sensitive to watershed alterations. Thesupply of elements to streams in all latitudesvaries with the geologic setting of the wa-tershed and with its plant community (e.g., Boltet al. 1984, Furch 1984, Lay and Ward 1987, Sioli1975, Stallard 1985, Stallard and Edmond 1983).The role of riparian vegetation in regulatingthe nutrient economy of streams deserves spe-cial study. Natural and anthropogenic distur-bance of the terrestrial ecosystem leads tochanges in the amount of nutrients enteringstreams (e.g., Correll et al. 1984, Dillon andKirchner 1975, Jordan 1987, Likens et al. 1970,Peterjohn and Correll 1984, Webster et al. 1983,Weller et al. 1986). These watershed-stream in-teractions have been documented in numerousreports, and this paper provides only a broadsummary.
A third reason is that within-stream processescan alter the timing, magnitude, and form ofelemental fluxes to downstream ecosystems(Meyer and Likens 1979, Rigler 1979), therebyinfluencing the availability of the element tobiological communities downstream in rivers,lakes, and estuaries.
Fourth, DOC dynamics are included here be-cause of the importance of DOC in stream en-ergy budgets. In most streams and rivers al-lochthonous inputs of carbon are importantenergy resources for the stream community.DOC is often a major form of allochthonouscarbon (e.g., Fisher and Likens 1973), but canalso be derived from autochthonous sources(Kaplan and Bott 1982). Allochthonous and au-tochthonous DOC fuel a microbial food web instreams (e.g., Edwards and Meyer 1987b, Locket al. 1984, Meyer et al. 1987a, 1987b, Rounicket al. 1982), and the importance of that foodweb relative to other energy flow pathwaysneeds careful study.
Fifth, many of the anthropogenic assaults onstreams have been nutrient additions leadingto major alterations of the stream community(e.g., Ball and Bahr 1974, Hynes 1969, Wuhr-
mann 1972). In addition, streams supply thewater needed by many communities and in-dustries, and the suitability of that water is oftendependent on the capacity of streams to removenutrients added as wastes upstream, or in otherwords, on the elemental cycling in the stream.
Controls on elemental supplyand availability
In this section we discuss factors that influ-ence the rate of supply of nutrients and organiccarbon to streams. We begin with processes op-erating at the level of the landscape and thenconsider the potential significance of short-termevents like storms in determining supply andavailability of elements. We then discuss therelative significance of longitudinal and laterallinkages as well as internal recycling as pro-cesses influencing elemental dynamics. We con-clude this section with a discussion of the spi-raling concept. The major processes consideredin our discussion of elemental dynamics are dia-grammed in Figure 1.
Throughout this discussion we emphasizefactors that control the availability of elementsto the biological community. It is critical thatstream ecologists move beyond lumping all or-ganic forms of an element together (e.g., as DOCor dissolved organic nitrogen [DON]), and rec-ognize that only a small portion of the com-pounds composing the DOC, DON, or dissolvedorganic phosphorus (DOP) pools in a systemare available to the biota. We should be consid-ering those compounds that will have the great-est influence on the biotic community. As ecol-ogists, we need to approach chemistry from theperspective of stream organisms and developways to characterize the bioavailability of com-pounds or classes of compounds. Organic formsof critical elements are the most abundant formin many streams. In Walker Branch, Tennessee,30-50% of dissolved P does not react with mo-lybdate and hence is assumed to be DOP (Segarset al. 1986), and 50% of the dissolved N exportis DON (Henderson and Harris 1975); in theKuparuk River, Alaska, 92% of dissolved N isDON (B. J. Peterson, unpublished data); DOPaccounts for 45-55% of the total P in streamsdraining forested watersheds in the U.S. (Om-ernik 1977). Despite their predominance, weknow little about the relative abundance of thoseorganic compounds that stimulate biological
412 J. L. MEYER ET AL. [Volume 7
REDUCTION / RELEASEGROUNDWATER
FIG. 1. Processes involved in elemental dynamics in a stream ecosystem.
growth, those involved in allelopathic inter-actions, and those which are recalcitrant or es-sentially inactive. This knowledge is crucial toan accurate assessment of elemental interac-tions with the biotic community.
Landscape processes and in-streamheterogeneity influence elementalsupply and availability
The major controls on element supply to astream include watershed geology and hydrol-ogy, soil processes, land-use practices, land-scape vegetation, and atmospheric loading.These factors vary both within and betweenlatitudes. Considerable information about theseinfluences exists for temperate streams, but lessis known about arctic or most tropical systems.For the Amazon basin, the effects of watershedgeology have been well-documented (e.g., Sioli1984). In the Arctic, dramatic (ten-fold) differ-ences in nitrate supply are observed in similar-sized tributaries draining different types of tun-dra ecosystems (B. J. Peterson, unpublisheddata). In North American streams, total phos-phorus content of streamwater is greater in wa-tersheds draining sedimentary rocks than inthose with granitic geology, and considerably
greater in streams in urbanized and agriculturalrather than forested watersheds (Dillon andKirchner 1975, Omernik 1977). Clearcuttingforests has resulted in increased concentrationsof nitrates in streamwater in a variety of wa-tersheds (e.g., Likens et al. 1970, Swank andDouglass 1975). Increased atmospheric loadingof hydrogen ions has resulted in decreased pHof streams in North America and Europe (Cosbyet al. 1985, Kramer et al. 1986, Mason and Seip1985). In highly weathered tropical soils, theinflux of elements from the atmosphere may bethe major net source of elements; the amountof elements reaching the stream will dependupon the utilization capabilities of the vegeta-tion (Kellman et al. 1982). Despite the potentialsignificance of atmospheric deposition of ele-ments to many tropical ecosystems, there are asyet few data with which to determine its im-portance.
These watershed- or landscape-level process-es define the overall supply of elements to astream and provide the framework within whichother processes operate on smaller spatial scalesand shorter temporal scales to regulate supplyand availability. A stream represents a spatiallydiverse nutrient environment, which is a re-flection of landscape processes, the patchy na-
1988] ELEMENTAL DYNAMICS IN STREAMS 413
ture of streambed habitats, and in-stream bio-logical activity. Numerous studies havedemonstrated changes in concentrations of nu-trients and DOC over short distances in streams(e.g., Johnson et al. 1981, Kaplan et al. 1980,Meyer and Tate 1983, Segars et al. 1986). Inaddition, patches of nutrient- or DOC-rich sed-iment may develop in response to localized con-centrations of organic matter or seepage ofgroundwater; algal and bacterial communitiesrespond to these (Crocker and Meyer 1987,Meyer et al. 1987b, Pringle 1985, Pringle andBowers 1984, Pringle et al. 1986). Hence it isinappropriate to consider only the nutrient en-vironment of the water when evaluating thenutrient supply of a stream.
A temporal perspective: the relativeimportance of short-term events
Much of the emphasis in studies of elementaldynamics in streams has been on those pro-cesses that release nutrients and DOC and sup-ply them to streams. Variation in rates of ele-mental supply to streams occurs on temporalscales ranging from hours to decades. We knowmost about seasonal and annual patterns andleast about longer term (e.g., decade) and dielchanges. Because stream chemistry reflects thelandscape which it drains, long-term changesin the landscape may result in changes in streamchemistry. On a geologic time scale, weatheringof a watershed is likely to alter the elementeconomy of drainage systems. On a shorter timeframe, anthropogenic impacts such as acidifi-cation of precipitation may also result in changesin elemental concentrations in streams. Suchchanges are difficult to detect owing to the needfor a long-term continuous data base. Long-termchanges in stream chemistry have been ob-served in New Hampshire; for example, mag-nesium concentration has decreased ~20% in20 years (Likens et al. 1985). Although the causeof this decrease is not certain, it is likely relatedto acidic deposition and associated changes inion exchange and/or weathering in the wa-tershed.
The rates of terrestrial processes such as rockweathering, elemental uptake by the vegeta-tion, and leaching of soil organic matter, andin-stream processes such as excretion of organ-ics by algae can vary seasonally leading tochanges in elemental supply rates. For example,in streams at Hubbard Brook Experimental For-
est, New Hampshire and Walker Branch, Ten-nessee, nitrate supply to the stream is greaterduring the winter because of the absence ofuptake by the terrestrial vegetation (Elwood andTurner 1988, Likens et al. 1977). In White ClayCreek, Pennsylvania, spring algal growth leadsto diel pulses of DOC (Kaplan and Bott 1982).The Caroni River, Venezuela, exhibits large sea-sonal fluctuations in DOC concentration (Pa-olini et al. 1983), whereas in the Caura River,Venezuela, DOC, total N, and total P concen-trations show little seasonal change despite ma-jor seasonal changes in discharge (Lewis 1986,Lewis et al. 1986). The proportion of the totalelement standing stock available to the biotaalso probably varies seasonally, but this is anarea where more work is needed. Examinationof changes in the proportion of element stand-ing stock available to the biota over a varietyof temporal scales will be a fruitful area forfuture research.
How important to the biota of streams are theless dramatic but longer-lasting seasonal shiftsin supply of elements relative to the more dra-matic but short-term changes seen during pulsedevents like storms? We don't have sufficient datato provide a good answer to that question. Inone sense, concentration at baseflow is whatsupports production, and hence slight seasonalchanges in concentration may have a major ef-fect. On the other hand, storms may serve as anelement subsidy by mobilizing previously un-available nutrient sources, and nutrients takenup during short periods of elevated concentra-tion may support growth over several days. Thedistinction between a disturbance and a subsidyis unclear. Increased flows of low magnitudebut short recurrence interval may provide el-emental inputs tthat subsidize the communitybetween major events. Larger storms with long-er recurrence intervals may be disturbances be-cause they scour the community and removemuch of the nutrient capital associated withbenthic organic matter. Are there thresholds forspates above which the short- and long-termsystem productivity is decreased, but belowwhich it is enhanced? This question of the roleof storms on the elemental dynamics of streamsis part of the larger group of questions on theimpact of disturbance on streams addressed byResh et al. (1988—see this issue). It is a questionrequiring further research.
Changes in element concentration withstream discharge have been documented in nu-
414 J. L. MEYER ET AL. [Volume 7
merous studies. In Table 1 we summarize dataon element relationships with discharge for 13constituents in 12 drainages in latitudes rang-ing from 68°N to the equator. There are no con-sistent changes in the relationships with lati-tude. One of the most striking features of thetable is the variation observed between ele-ments and streams. In most streams NH4 con-centration does not change with discharge,whereas DOC, particulate organic carbon (POC),total phosphorus (TP), and suspended sedi-ments consistently increase. The increases inPOC, TP, and suspended sediments are a con-sequence of increased ability to erode and tokeep particulate matter suspended at higherflows. Depending on the stream being sampled,soluble reactive phosphorus (SRP), Mg, Ca, SO4,and NO, concentrations either increase, de-crease, or remain constant as discharge changes.In two streams, the relationship between NO3
concentration and discharge varies betweenstorms; Na and K also show between-storm vari-ability. In general, Na concentration and pHeither are constant or decrease with increasingdischarge. Potassium concentration is eitherconstant or positively related to discharge.
The variability in relationships between con-centration and discharge that is clearly shownin Table 1 is to be expected because data col-lected from several streams in the same basinvary considerably (D. L. Correll, SmithsonianEnvironmental Research Center, personal com-munication). Within the Rhode River wa-tershed the pattern changes with magnitude ofthe storm, season, year, and even time since lastrainfall. Adjacent watersheds exhibit differentpatterns based on differences in topography orland use. Even the method of .sampling (e.g.,grab vs. flow-weighted) alters the apparent pat-tern. Thus, although we may see general pat-terns emerging within data sets, the variationis large. Some of the variation is also associatedwith the different flow paths of water throughthe watershed. For example, in Walker Branchcalcium and magnesium concentrations declinewith increasing discharge because of dilutionof groundwater that has been in contact withthe dolomitic bedrock by quick-flow soil waterthat has not. Sulfate increases as a result of theinput of soil water flowing through the uppersoil horizons where the sulfate adsorption ca-pacity is low, sulfate retention is minimal, andwater soluble sulfate levels in soil are high (El-
wood and Turner 1988). Although our data baseon variation of elemental concentration withdischarge is extensive, we know little about howchanges in discharge affect concentrations ofbiologically available forms of critical elements.For example, although the concentration of DOCincreases with discharge in 11 of the 12 streamslisted in Table 1, we do not know if all formsof DOC increase or if, for example, the refrac-tory forms increase to a greater extent than themore labile ones. In Bear Brook, New Hamp-shire, carbohydrate concentration did not in-crease as rapidly as total DOC with discharge,suggesting dilution of labile DOC during storms(McDowell and Likens 1988). We need to clarifythe relationship between the pool of biologi-cally available elements, the total element pool,and the elements washed downstream duringstorms. This area of stream ecology will likelybenefit from the application of new analyticalmethods, such as measuring specific forms ofDOC and speciation of key elements. To un-derstand linkages between elements, biota, andsuspended particles, it will be important to as-sess how the biological availability of elementsvaries with stream discharge.
We also know little about the effect of storm-induced fluxes of elements on the biologicalcommunity. Do storm-generated nutrient pulsesinitiate important processes such as algal re-production? Are storms a nutrient subsidy forthe community or a nutrient loss? The answerprobably varies with both the intensity of thestorm and the season in which it occurs. A lowmagnitude storm will probably not scour thestreambed, but can supply or mobilize nutrientsin the channel; a high magnitude storm canerode nutrient-depleted surface layers of rockor organic matter, but may also destroy most ofthe biotic community. For example, in WhiteClay Creek, Pennsylvania, a light rain on fresh-ly fallen leaves increased DOC concentrationfrom <2 to >5 mg C/L, stimulating bacterialproduction; but heavy rains scour the streambedcommunity (T. L. Bott and L. A. Kaplan, un-published data). The exchange of materials be-tween the channel and interstitial zone is oftencontrolled by the volume of water in the chan-nel (i.e., the pressure head); hence at higherflows, the flux of materials to the hyporheoswill probably increase. A small storm may havelittle impact on the hyporheic zone, a moderatestorm may pump more material into it, but an
TABLE 1. The relationship between element concentration and stream discharge in a variety of streams over a range of latitudes. A " + " indicates thatelement concentration increases with increasing discharge, "-" indicates that concentration decreases, and "0" indicates that there is no change in concentrationwith discharge. More than one symbol is used when the relationship between concentration and discharge varies between storms.
fft• Data set from grab samples during storms.b Asymptotic above 1.1 m3/s.
1. Peterson et al. 1983, B. J. Peterson et al., unpublished data.2. Fisher and Likens 1973, Johnson et al. 1969, Likens et al. 1967, McDowell and Wood 1984, Meyer and Likens 1979.3. Lewis and Grant 1979.4. Bott et al. 1984, Kaplan 1980, Minshall et al. 1985, R. L. Vannote—Stroud Research Center—unpublished data.5. Correll et al. 1987, D. L. Correll—Smithsonian Environmental Research Center—unpublished data.6. Elwood and Turner 1988.7. S. W. Golladay and J. R. Webster—Virginia Technical Institute—unpublished data; Gurtz et al. 1980, Meyer and Tate 1983.8. McDowell 1984, W. H. McDowell, unpublished data.9. Lesack et al. 1984.
10. C. Ishizaki, I. Marti, and R. Colmenares, unpublished data.11. Gaudetand Melack 1982.
416 J. L. MEYER ET AL. [Volume 7
eroding storm will wash out pockets of organicmatter and nutrients that have accumulatedsince the last storm in protected sediment en-vironments. The impact of storms on the sedi-ment community will depend on the depth ofscour relative to the extent of the hyporheos.
The effect of high discharges on element dy-namics will vary depending on the major path-ways for element uptake in the stream. For ex-ample, the uptake length for phosphorus inWalker Branch varies seasonally with the stand-ing stock of coarse particulate organic matter,and generally increases after storms owing tothe loss of organic matter from the stream bedduring the storm (Mulholland et al. 1985b). Thisorganic matter is the dominant sink for dis-solved phosphorus in this stream. Uptake lengthis not very .responsive to changes in epilithicalgal standing stock in Walker Branch; the mass-specific uptake rate of epilithic algae is high,but their biomass is low (Newbold et al. 1983).In an unshaded stream where the algal com-ponent of phosphorus uptake may be more im-portant, the algal response to storms should reg-ulate changes in nutrient dynamics. We needmore information on these types of storm effectfor a range of elements and streams.
The effect of storms will also vary dependingon the relative importance of particulate anddissolved forms of the element of interest. Par-ticle concentration increases during storms inmost streams (Table 1). Hence an element likephosphorus, which is found primarily in theparticulate phase, will be lost disproportionate-ly during storms (e.g., Meyer and Likens 1979).Storms may also have important consequencesfor particle-associated microbiota. For example,in the Rhode River, most bacteria in the watercolumn are not attached to particles (Rublee etal. 1984) and about three orders of magnitudelower in concentration than in the underlyingsilty sediments where bacteria are mostly at-tached to particles (Rublee et al. 1983). Whensediment resuspension occurs during storms,significant bacterial biomass and activity arethen associated with particles (Rublee et al.1984). Similar changes in particle-bound bac-teria have also been observed in headwaterstreams (Palumbo et al. 1987). In contrast, in asubtropical blackwater river with an extensivefloodplain, the proportion of particle-associatedbacteria remains small even during floods, pre-sumably because large numbers of unattachedand inactive bacteria are washed from flood-
plain soils during inundation (Edwards 1987,Edwards and Meyer 1986). Thus in streamswithout extensive floodplains, storms may in-ject particle-bound and metabolically activebacteria into the water column and subsequent-ly displace them downstream or wash them outof the system.
Geomorphology has a major impact on par-ticle transport and hence also on elemental dy-namics in streams. Retention devices like debrisdams are critical in retaining particulate organicmatter in streams (e.g., Bilby and Likens 1980),and the amount of organic matter is a key de-terminant of the stream's ability to remove dis-solved elements from solution (Mulholland etal. 1985b). But are these sites of organic matterstorage also major sites of nutrient regenera-tion? And how do geomorphic features such asgradient or riffle:pool ratio influence particletransport and hence organic matter storage andthe capacity for retention of dissolved elementsin flowing waters? These questions are as yetunanswered.
The effects of storms on elemental cyclingwill also be influenced by the season in whichthey occur and by the extent and nature of thefloodplain. Floodplains can increase the reten-tiveness of the system for dissolved and partic-ulate nutrients as a consequence of the largerstorage volume, more circuitous routing, andgreater biotic and abiotic sorption than occursin the stream itself.
Although many questions still remain aboutthe influence of increases in discharge on ele-mental dynamics, we know considerably moreabout the influence of these events than aboutthe influence of drought. Studies on intermit-tent streams have shown that drying will havea profound impact on the insect community.Immediately after rewetting, the invertebratecommunity of a seasonally dry stream is differ-ent from that of the perennial stream, but con-verges to that of the perennial stream with time(Delucchi 1988). We know little about the im-pact of drying, rewetting, and changes in theinvertebrate community on the nutrient econ-omy of these streams; yet a periodic droughtmay be a greater disturbance than a spate.
A spatial perspective: longitudinal vs.lateral controls on elemental dynamics
A central tenet of the River Continuum Con-cept is that community structure and function
1988] ELEMENTAL DYNAMICS IN STREAMS 417
are dominated by upstream processing, i.e., lon-gitudinal linkages in the ecosystem are strong(Minshall et al. 1983, Vannote et al. 1980). How-ever, the relative importance of lateral influ-ences (floodplain or riparian zone), upstreamlinkages, and internal recycling (e.g., withinbiofilms or between deep and surface sedi-ments) have not been sufficiently well exam-ined to provide a real test of the concept. Forexample, does retention of elements in up-stream reaches lead to nutrient limitationdownstream? If lateral inputs from the riparianzone or regeneration of nutrients within thereach are important sources of biologicallyavailable nutrients at a point in the stream, thenthe retention of nutrients upstream will haveonly a minor impact. In that case the structureof the upstream community will have little in-fluence on processes occurring downstream.
Strong longitudinal linkages have been dem-onstrated in some streams. For example, in Syc-amore Creek, Arizona, uptake of nitrogen up-stream leads to blue-green algae dominating thealgal community downstream (Fisher et al. 1982).In Walker Branch, SRP concentration declinesdownstream from major groundwater seeps be-cause of uptake or conversion of SRP withinthe stream channel in fall and winter (Segarset al. 1986). Hence the supply of P to down-stream communities is reduced by the uptakeof P in upstream reaches. If the P taken up bydetritus upstream is regenerated into the water,then retention may act as a stabilizing influenceon the P supply. However, if the P taken up bythe detritus is washed out or released back tothe water in a less available form, it representsa permanent sink of P for upstream commu-nities and a potential source of P to downstreamcommunities if it is converted to an availableform downstream. The temporal and spatialscales at which one considers the question ofelemental uptake, storage, and transport thusbecome important. The relative importance oflateral vs. longitudinal vs. internal controls onavailable nutrients will also vary temporally.For example, under drought conditions, inter-nal recycling of elements in a reach would pre-sumably become a more important source ofavailable nutrients than upstream sources.
The regeneration /remineralization of nu-trients in streams is an area in need of process-level work to determine the rate of regenera-tion, the forms of regenerated nutrients, andtheir biological availability. For example, ap-
proximately half the phosphorus taken up byleaf detritus in a heterotrophic laboratory streamchannel was retained in the detritus and halfwas lost either by regeneration to soluble P,grazing, or sloughing (Elwood et al. 1988). Theeffect of these processes on the supply of P todownstream systems depends on the biologicalavailability of the regenerated P and the per-manence with which P is retained by detritus.Some regenerated P appears to be less availableto microbes than orthophosphate (Mulhollandet al. 1988), thereby reducing the supply ofavailable P to downstream communities. Thisneeds to be examined in a greater variety ofstreams and in relation to size spectra and com-position of inorganic particles.
Clearly lateral input from natural or anthro-pogenic sources can be important for lotic com-munities at all scales. The potential for in-creased importance of lateral inputs anddecreased importance of longitudinal linkagesis great in rivers with extensive floodplains.Fisheries workers have documented a positiverelationship between fish production and flood-ing intensity both within and between rivers(e.g., Moses 1987, Welcomme 1985). Less isknown of the impact of floodplain extent andnature on the nutrient economy of streams. Twolines of evidence suggest that the Amazonfloodplain is an important source of the carbonto the river: its tributaries do not provide ade-quate organic matter to support observed levelsof respiration in the river (Sedell and Richey1988), and measures of floodplain production,respiration, and decomposition suggest that 11%of carbon fixed annually is exported to the river(Junk 1985). In general, watersheds with asso-ciated floodplains and wetland export more or-ganic matter than drainages without wetlands(Schlesinger and Melack 1981), attesting to theimportance of lateral inputs in these systems.Riparian wetlands will also influence the nu-trient economy of the fluvial ecosystem throughtheir retention of nitrogen and phosphorus(Mitch and Gosselink 1986).
In rivers with extensive floodplains a greatersurface area is exposed, and hence atmosphericinteractions are probably of greater importance.Methanogenesis is very active in tropical flood-plains such as those bordering the Amazon, andefflux of methane from these wetlands makes anotable contribution to tropospheric methane(Bartlett et al. 1988). Processes such as nitrogenfixation and denitrification can greatly influ-
418 J. L. MEYER ET AL. [Volume 7
ence the nitrogen economy of all flowing waters,yet we have little comparative data from a di-versity of systems on rates of these processes.In Oregon headwater streams, N fixation ac-counted for only 5% of total N inputs (Triska etal. 1984), but in an unshaded California stream,high rates of N fixation were measured (Home1975, Home and Carmiggelt 1975). In Okla-homa streams grazing activities of fishes leadto an algal community with abundant blue-greens, and hence the potential for high ratesof nitrogen fixation (Power et al. 1985). Deni-trification was found to be a significant nitrogensink in southern Appalachian streams (Swankand Caskey 1982) and in a nitrate-enriched Ca-nadian river (Hill 1979, 1983). These processesvary considerably within a single stream; hencewe anticipate variation in the rates and impor-tance of processes like nitrogen fixation anddenitrification will be as great within as be-tween latitudes.
Elemental dynamics in a stream may also beinfluenced by the riparian vegetation (Peter-john and Correll 1984). The riparian forest re-moved 82% of dissolved N, 54% of dissolved P,and 42% of dissolved Ca draining into a GeorgiaCoastal Plain stream with an agricultural wa-tershed (Todd et al. 1983). Trees with nitrogen-fixing symbionts are a common feature of manyriparian zones; in southern Appalachian wa-tersheds dominated by the nitrogen-fixing blacklocust (Robinia pseudoacacia), nitrate concentra-tions are higher than in adjacent reference wa-tersheds, and leaf decomposition rates are ac-celerated by this additional nitrogen (Meyer andJohnson 1983). The strength of these types oflateral linkages needs to be assessed with re-spect to the longitudinal linkages discussed ear-lier.
Recycling of elements within a stream reach
The relative importance of external vs. inter-nal sources of elements (regenerated nutrients)needs to be assessed in terms of the biotic com-munity. How much of the biological demandfor nutrients is met by recycling vs. externalloading? This question has been central to manystudies of nutrient dynamics in lake and marinesystems (e.g., Eppley and Peterson 1979), buthas not been adequately addressed in streams.The answer will vary depending on the elementbeing considered and the stream. It may also
vary temporally and with position in the streamnetwork.
The extent of biotic regulation of elementalsupply and availability needs to be comparedwith the impact of factors such as storms andseasonal changes in element supply rates. Forexample, how does the impact of grazers on thenutrient economy of a stream compare with theimpact of the ten-year flood? Assessments of theimpact of biotic and abiotic controls on streamelemental dynamics are critically needed instreams spanning a range of latitude and dis-turbance regimes.
The active and diverse community of mac-rofaunal, meiofaunal, and microfaunal con-sumers is likely to have a significant effect onnutrient availability in streams. This is a subjectin need of further attention. For example, instreams where relatively large consumers likefish or snails are abundant, their feces may proveto be a key link in nutrient cycles (W. J. Mat-thews, University of Oklahoma, personal com-munication). Meio- and microfaunal contribu-tions to stream nutrient dynamics will be animportant area for further research. These or-ganisms have been demonstrated to be impor-tant remineralizers in other ecosystems (Bars-date et al. 1974, Fenchel 1970, Johannes 1965);their importance in sediment-dominated sys-tems like streams needs to be assessed. It is like-ly that the importance of faunal remineraliza-tion will vary with the nature of the sediments,particularly grain size, because of its impact onthe composition of the fauna. In other aquaticsystems, microflagellates predominate in fine-grained sediments, ciliates in coarser-grainedsediments, and nematodes in sediments withhigh silt and clay content (Fenchel 1978).
The microbial food loop (sensu Pomeroy 1974)needs to be examined in sediment-dominatedsystems like streams. In some pelagic systemsthe microbial loop appears to be a sink for car-bon (Ducklow et al. 1986, but see also Sherr etal. 1987), but a regenerator of other nutrients(Goldman and Caron 1985). In sediments, thereverse may be true. In streams most of thebacterial biomass is found in the sediments, evenin larger rivers (Meyer et al. 1987b). Where thebacteria are particle-bound, the loop may be amore direct link in the food web because largerconsumers can ingest small microbes as a "sand-wich" of some biomass and a lot of detritus;hence microbes are available to larger organ-
1988] ELEMENTAL DYNAMICS IN STREAMS 419
isms with fewer trophic transfers. For example,bacteria on decomposing leaves are directlyavailable to leaf-shredding insects, althoughtheir biomass does not appear to be a majorcarbon source for the insects (Findlay et al. 1986).Bacterivorous meiofauna such as copepods areabundant in stream sediments (O'Doherty 1985),and these copepods are found in the guts ofmacroinvertebrates like dragonflies (Wallace etal. 1987). In addition, many stream-dwelling,filter-feeding macrofauna (e.g., black flies) havethe capacity to remove bacteria-sized particlesfrom the water (e.g., Edwards and Meyer 1987b,Wotton 1980). Hence the bacteria are convertedinto larger morsels of food with few trophictransfers and therefore greater efficiency. Thenutrient regenerating function of the loop willbe influenced by the nature of the sediment, inparticular its ability to sequester nutrients. Nu-trients regenerated in the sediments are vul-nerable to abiotic sorption processes, and hencea smaller portion may be available to supplybiotic needs than in the water column. Theseideas need to be pursued.
Critical interfaces for the control of elementaldynamics in streams appear to be floodplainsor riparian zones, biofilms (e.g., epilithon), andthe sediment-water interface. Because of sharpgradients in physical and chemical parameters(e.g., pH, Eh, oxygen) that can exist at theseinterfaces, they can be exceedingly importantin regulating elemental flux. Work needs to bedone in these ecotones (cf. Naiman et al. 1988and Pringle et al. 1988—both in this issue). Forexample, the relative importance of sedimentand overlying waters as sources of elements forepilithic and epipsammic biofilm communitieshas been examined (Bott et al. 1984, Pringle1987, Pringle and Bowers 1984) and deservesfurther attention. The intimate relationshipsbetween primary producers, heterotrophic mi-croorganisms, and consumers within the auf-wuchs suggest that within-biofilm interactionsas well as biofilm-water column exchanges willhave an impact on elemental dynamics instreams.
The hyporheos constitutes an important res-ervoir of nutrients and organic matter, and itsrole in the nutrient economy of streams is wor-thy of further investigation. The hyporheos isa patchy environment, as discussed earlier; vari-ation in nutrient and DOC concentration, oxy-gen content, and flow regime is considerable
(Hynes 1983, Rutherford and Hynes 1987). Theabundant organisms in this habitat are sup-ported by allochthonous organic matter pro-cessed by an active biofilm community of bac-teria and fungi rather than primary producers(e.g., Bretschko and Leichtfried 1988; J. A. Stan-ford, University of Montana, unpublished data);these organisms will probably be more signif-icant than algae in nutrient dynamics in thiszone. The hyporheic biofilm community ap-pears to be surface-area limited (Leichtfried1985). Lateral inputs of dissolved substancesfrom the watershed or riparian zone frequentlypass through the hyporheic zone; in some in-stances DOC seems to be largely consumed inthe hyporheos before it reaches the overlyingwater (Wallis et al. 1981). Exchanges of dis-solved and particulate substances between sur-face water and hyporheos are probably dis-charge dependent, although this is a subject onwhich more research is needed. Metabolic ac-tivity is considerable in the deep sediments ofsome streams (e.g., Grimm and Fisher 1984,Meyer 1988), and hence one would expect thatbiological activity in the hyporheos would havea profound influence on elemental dynamics inthe stream as a whole. Our data in this area aresorely lacking.
The potential importance of anaerobic pro-cesses in stream ecosystems is beginning to berecognized on spatial scales ranging from theinterior of a decomposing leaf (D: Lawson,Michigan State University, personal commu-nication) to microzones in the sediments (Dahmet al'. 1987, Jorgensen and Revsbech 1985) to thebottom sediments of beaver ponds (Naiman etal. 1986). In general, studies of anaerobic pro-cesses in streams consider pockets of anaerobicconditions in predominantly aerobic systems,for example denitrification occurring in the epi-lithic film of a well-oxygenated stream (Triskaand Oremland 1981, Ventullo and Rowe 1982).Anaerobic processes could be important in re-generating nutrients and could influence thedynamics of nitrogen, phosphorus, sulfur, car-bon, and iron in streams. The importance ofanaerobic processes in the overall elementeconomy of a stream will vary with disturbancefrequency, extent of debris accumulations, andnature of the benthic substratum. In streamswhere organic loading is high and sulfate andnitrate concentrations are low, one would ex-pect methanogenesis to be an important carbon
420 J. L. MEYER ET AL. [Volume 7
sink. Anaerobic processes occurring in the wa-tershed, particularly in the riparian zone andfloodplain, will influence the supply rate of ele-ments to the stream.
Interactions between elements have beendemonstrated to influence element concentra-tions in streams. For example in streams in theGreat Smoky Mountains, pH and alkalinity in-crease downstream owing to tributary inputs,and the concentration of soluble monomericaluminum decreases. Dilution from tributarieswith low aluminum concentration does not ac-count for the entire aluminum decrease, indi-cating that aluminum is precipitating and/orbeing adsorbed in the stream channel under thehigher pH conditions. The downstream de-creases in P and DOC are most likely due totheir adsorption to precipitated hydrous alu-minum oxides (J. W. Elwood, unpublished data).The consequence is an oligotrophication ofdownstream reaches. These kinds of elementalinteractions are probably important in otherstreams as well and are worthy of investigation.Aquatic humic substances interact with metalsand nutrients, and in streams where they arehigh in concentration (e.g., blackwater rivers),complexation of humics with metals and ni-trogenous compounds can influence the avail-ability of metals and nutrients to the biota(Thurman 1985).
Humic substances may also be important inbiotic interactions in streams. The high DOC inblackwater rivers may include a bioactive sub-stance that can influence macroinvertebrate dis-tribution and behavior through its impact onolfaction. Olfaction influences the behavior ofthe organisms in which it has been studied (e.g.,mayflies [Peckarsky 1980], stoneflies [Williams1986], and other groups). Grazing insects areable to taste and appear to use this cue whenfeeding (Hart 1981, Kohler 1984); high concen-trations of humic substances may alter olfactionand therefore feeding behavior.
Spiraling (Elwood et al. 1983, Newbold et al.1981) is a unifying concept that is applicable tostudies of elemental dynamics in streams. Spi-raling is nutrient cycling combined with down-stream transport, i.e., nutrients are displaceddownstream as they pass through a cycle. Thecycle is both open and closed. It is closed in the
sense that a nutrient can pass through the samechemical or trophic level many times during itsresidence in a stream. It is open in the sensethat completion of the cycle does not occur inplace, but rather involves some downstreamdisplacement before the cycle is closed. The spi-raling concept thus deemphasizes the view ofstreams as spatially bounded units of the land-scape because cycling continues indefinitely asnutrients move along a spatial (i.e., longitudi-nal) axis.
The spiraling concept offers a useful tool forlatitudinal comparisons of elemental dynamicsin streams and for assessing the importance ofinternal recycling vs. lateral and longitudinallinkages. For example, if one were interested inassessing controls on community structure andproductivity at a point in a stream, a measureof spiraling length would permit an evaluationof the influence from upstream at that point; alonger spiraling length implies more inputs andhence greater controls from upstream reaches.In this sense, a study of spiraling broadens ourfocus from a single spot in the stream to reachesand how they are linked. The spiraling conceptalso offers a mechanism for linking studies ofcommunity dynamics and element cycling instreams; for example, measures of nutrient spi-raling after experimental manipulations ofgrazers allow one to assess the impact of grazerson nutrient dynamics (e.g., Mulholland et al.1983). Measures of spiraling are most interest-ing for the limiting nutrient in a stream.
Spiraling length can be measured with stableisotopes. For organic carbon, all that is requiredis a measure of downstream flux and respiration(Newbold et al. 1982a). For other elements, up-take length can be measured with additions ofthe element of interest, although it is criticalthat the additions be at concentrations belowthe half-saturation constant for uptake (New-bold et al. 1982b). In this type of experiment,turnover length would have to be measured byradioactively labelling a compartment and fol-lowing the loss of radioactivity, but this mea-sure could be done reliably in laboratory chan-nels. It may be possible to use dilution ofexperimentally added stable isotopes as a mea-sure of regeneration; for example the 15N/14Nratio will change as 14N is regenerated along astream reach. This technique has been appliedto planktonic communities (Morrissey andFisher 1988). The measurement of turnover
1988] ELEMENTAL DYNAMICS IN STREAMS 421
length would benefit from further methodo-logical development.
An important future research direction forthe measurement of carbon spiraling in streamswill be to differentiate the components of totalorganic carbon spiraling in a river. Initially thatwill probably mean considering DOC and POCseparately; that is straightforward for the mea-sure of transport, but more problematic for themeasure of respiration. The next step will be toimprove analytical methods to dissect DOC intoits components and to compare spiraling of frac-tions that differ in their bioavailability. Ideallywe wish to measure spiraling length of biolog-ically available DOC fractions, but this is notpossible at the moment.
Carbon spiraling length has been measuredin a variety of streams (e.g., Edwards and Meyer1987a,Minshalletal. 1983,Newboldetal. 1982a),but spiraling length for a nutrient has only beenmeasured for phosphorus in Walker Branch,Tennessee (Mulholland et al. 1985b, Newboldet al. 1981,1983) and in experimental laboratorychannels (Mulholland et al. 1983,1985a). It needsto be done for other elements in other types ofstreams. In Walker Branch, spiraling length var-ied from 10 to 160 m, largely due to differencesin standing stock of organic matter (Mulhollandet al. 1985b). Other factors that will influencespiraling length are hydrologic regime, grazeractivity, temperature, and algal activity. In largerivers the spiraling concept may be most use-fully applied by dissecting the system and ex-amining the role of the various components incontrolling spiraling length.
One of the critical questions to which thespiraling concept can be applied is the extentto which the biotic community retards down-stream transport of elements, i.e., the relativeimportance of biotic retention vs. abiotic trans-port. Evidence from Walker Branch shows thatthe biotic community associated with leaf de-tritus is responsible for most of the uptake ofphosphorus (Mulholland et al. 1985b); macroin-vertebrates do not account for much of the spi-raling length of P because a relatively smallfraction of the total spiraling flux of P passesthrough the invertebrate community (Newboldet al. 1983). Invertebrates, however, may alterthe spiraling indirectly through activities suchas grazing. An application of the spiraling con-cept that should provide insight into mecha-nisms controlling nutrient dynamics in streams
is the impact of consumer (either fish or inver-tebrate) density or species diversity on spiralinglength. These studies of the impact of structureon function would be of broad ecological in-terest.
Stream processes regulated by nutrientsupply and availability
The discussion to this point has consideredcontrols on nutrient supply and availability instreams; we now change our perspective andconsider what processes are controlled by nu-trient supply and availability.
Productivity and decomposition
In well-lit streams, primary productivity isoften stimulated by additions of nitrogen orphosphorus (e.g., Grimm and Fisher 1986), al-though fewer data are available for tropicalstreams (e.g., Pringle et al. 1986). It is likely thatthe extent and nature of nutrient limitation willvary as much within as between latitudes. Thesubject of nutrient limitation of primary pro-ductivity in streams has been extensively re-viewed (e.g., Bott 1983). An important area forfurther research is the interaction between nu-trient limitation and grazing pressure as con-trols on algal productivity. A fruitful experi-mental approach would be to combine nutrientlevels and grazing levels to assess their relativeimportance. These ideas are developed belowin the section on whole system manipulations.
Concentrations of micro-nutrients (Patrick1978), macro-nutrients (Pringle and Bowers1984), and grazing (Gregory 1983, Patrick 1970)can also affect algal species composition. A use-ful extension of this information will be to con-sider how changes in algal community structureinfluence rates of nutrient recycling in thestream. For example, a shift to more palatablespecies may accelerate nutrient regenerationthrough grazing.
Many factors influence the rate at which leaflitter decomposes in streams, including nu-trient concentration in the water (reviewed byWebster and Benfield in 1986). Some studieshave shown nitrogen to be the critical element,some phosphorus, and some have provided noevidence for nutrient limitation (Webster andBenfield 1986). If the autotrophic community ofa stream is nutrient limited, it does not neces-
422 J. L. MEYER ET AL. [Volume 7
sarily follow that the decomposer communitywill be. For example, in the Kuparuk River,Alaska, phosphorus additions stimulate pri-mary productivity, but not Carex decay (Peter-son et al. 1985). What is needed at this point isa clear assessment of the conditions under whichnutrients will be limiting, and what factors in-fluence which will be the limiting nutrient. Bothmicrobial oxidation and mass loss due to bioticand abiotic fragmentation occur during leaf de-cay. In addition to watershed influences onstream chemistry, the relative importance ofthese two processes (among both leaf speciesand streams) will probably influence the degreeto which decay is nutrient-limited; where mi-crobial processes are more important, the po-tential for nutrient limitation should be greater.
In studies of leaf decomposition it is criticalthat we move beyond simply assessing mass lossand attempt to separate losses due to microbialprocesses, invertebrate activity, and physicalfragmentation. How does the fractional loss frommicrobial respiration vary with leaf species (e.g.,greater for "fast" species?) and what are theimplications of this for the stream ecosystem?For example, if fragmentation is the primarymechanism of mass loss, some of the FPOM andassociated nutrients remain in the stream; how-ever if microbial respiration is dominant, thecarbon is lost and nutrients are regenerated andmade immediately available.
Throughout most of this discussion, we haveindicated that we anticipate variation withinlatitudes to be as great as variation betweenlatitudes. Leaf decomposition may be an excep-tion to this. We doubt that there will be a clearlatitudinal gradient in the degree to which nu-trients limit decomposition, but consistent dif-ferences in other controls on leaf decay rate areexpected. Although a range of temperature con-ditions exist in tropical streams (Covich 1988—see this issue), there are more warmwater en-vironments in lower latitudes, and elevatedtemperature clearly increases decay rate (Web-ster and Benfield 1986). The timing of leaf inputis also different—pulsed in higher latitudes andmore continuous in lower latitudes. Clear dif-ferences in chemical composition of leaves alonga latitudinal gradient could also influence decayrate. There are well-documented differences indecay rate of different leaf species (Webster andBenfield 1986). To the extent that allelochemi-cals in tropical leaves are antifungal and anti-
bacterial, they will retard decay rate unless thestream microflora have evolved a resistance tothese compounds. Allelochemicals that protectthe plant from terrestrial herbivores may alsoinfluence their palatability to shredders andhence their decay rate. Potential trace metal in-fluences need to be considered in evaluatinglatitudinal gradients in leaf decay. Althoughsome tropical plants appear to be enriched inaluminum (Sanhueza et al. 1988), some tem-perate riparian species (e.g.. Rhododendron) arealso aluminum accumulators; the relevant ques-tion is the relative abundance of aluminum (ora similar substance) accumulators along a lati-tudinal gradient. Temperate leaves placed intropical streams decompose more rapidly thanin a temperate stream (J. Stout, Michigan StateUniversity, personal communication). Morework with reciprocal transplants would behelpful in clarifying the various factors (e.g.,environmental conditions, leaf species, adap-tations of stream microflora) leading to tem-perate vs. tropical differences in leaf decay rate.
Little is understood about the relative im-portance of bacteria and fungi in leaf decom-position along a latitudinal gradient in streamsor about the extent to which nutrient regimecontrols their relative importance. There willclearly be changes in fungal species composi-tion in response to changing temperatures(Suberkropp 1984), but the influence of suchchanges on leaf decay rate needs to be exam-ined.
Detrital decay rate is greatly influenced bythe size of the decomposing organic matter.Fungi will be more abundant on large particles,and their role in decomposition relative to bac-teria will vary accordingly. Hence the relativesignificance of nutrient limitation may also varywith particle size. This idea can be extendedalong the entire size spectrum: wood, wholeleaves, fine particulate organic matter (FPOM),dissolved organic matter (DOM). Measures ofdecay rates of coarse particulate organic matterin streams abound in the literature (Websterand Benfield 1986); we know much less aboutthe decay (i.e., conversion to CO2) of FPOM orDOM in streams. The extent to which nutrientsupply rates influence rates of degradation ofDOM and FPOM in streams is an important areafor future research because DOM frequentlyprovides a major fraction of allochthonous or-ganic matter inputs (e.g.. Fisher and Likens
1988] ELEMENTAL DYNAMICS IN STREAMS 423
1973), and FPOM is frequently the dominantcomponent of organic matter standing stock instreams (e.g., Minshall et al. 1983).
Whole system manipulations in studiesof nutrient limitation
There is potential for important research inexamining the interaction between nutrientlimitation and other community processes instreams. Bioassay techniques for determiningnutrient limitation vary considerably, and in-novative approaches are needed. Whole systemexperimental manipulations offer a useful ex-perimental approach that is particularly appro-priate for testing one's conceptual understand-ing of mechanisms controlling structure andfunction of stream communities. The wholestream experimental approach allows one toquickly demonstrate which components of theentire system are most responsive to the ma-nipulated factor. Subsequent follow-up exper-iments can focus on the mechanisms of re-sponse and the controls. Without the wholesystem perspective, considerable effort could bemisdirected to examine interactions that are oflesser importance in the stream. Experimentalmanipulations of major components of the com-munity are necessary before we begin analyz-ing the influence of finer-scale changes in com-munity structure. However, we must exercisecaution in drawing inferences from wholestream manipulations that go beyond theboundaries of what the system experiences nat-urally. The value of whole stream manipula-tions is evident when one considers questionslike the importance of cascading trophic effects(Carpenter et al. 1985) in streams.
Considerable evidence exists that cascadingtrophic effects are important in streams. Studieshave demonstrated that invertebrate predatorscan control prey populations (reviewed byPeckarsky in 1982,1984), and that grazers exerta strong influence on algal community com-position, biomass, and productivity (reviewedin Gregory 1983). There is, however, little evi-dence that fish control invertebrate communitycomposition or standing stock (reviewed by Al-lan in 1983).
Studies of cascading trophic effects in streamsneed to be extended along the entire food web,from consumers to nutrients. In streams where
this has been done, whole system experimentalmanipulations provided insight. Phosphorusadditions to a tundra river lead to increasedprimary productivity, increased biomass ofBrachycentrus and Baetis, decreased biomass ofSimulium, and increased fish production (Peter-son et al. 1985), clearly demonstrating the in-fluence of nutrient limitation at several trophiclevels. Phosphorus enrichment of a southernAppalachian stream increased leaf decomposi-tion rate, standing stock of blue-green algae,and grazer density; because of increased graz-ing pressure an initial increase in total algalabundance was not sustained (Elwood et al.1981). In an Oklahoma stream, experimentalmanipulation demonstrated that piscivorous fishdecreased the abundance of dominant herbiv-orous fishes, and their decline resulted in in-creased epilithon standing stocks and lower nu-trient concentrations—a clear demonstration ofcascading trophic effects (Power et al. 1985). Ira southern Appalachian stream, removal of ma-croinvertebrates decreased leaf processing rates,greatly reduced seston output, and lead to in-creases in meiofaunal standing stock (Wallaceet al. 1982).
The possibilities are numerous for significantresearch in examining the interaction betweenfood web components extending from nu-trients to fish. For example, does nutrient levelaffect the influence that predators have onstream community structure through controlson population levels of grazers? Conversely, dopredators, through their impact on primaryconsumers, alter nutrient dynamics in the sameway that they can alter rates of litter decom-position (Oberndorfer et al. 1984)? This kind ofresearch has provided insight into regulationof lentic community structure (e.g., Neill 1981,Neill and Peacock 1979), and should proveequally useful for studies of stream communi-ties.
In this paper we have identified areas wherewe think there are critical gaps in our under-standing of elemental dynamics in streams. Inthis section we suggest possibilities for newtypes of research programs in stream ecologythat if implemented would promote the type ofresearch that would fill these critical gaps.
424 J. L. MEYER ET AL. [Volume 7
Development of methods
Although progress in stream ecology is notnecessarily limited by available methods, ap-plication of methods from other fields to prob-lems in stream ecology should lead to new in-sights. One mechanism to foster this would besponsoring workshops aimed at cross-discipli-nary application of methods to a question instream ecology. Follow-up funds would be nec-essary to try out the concepts developed at theworkshop. In many cases instruments and pro-cedures have been designed for clean, puresamples at high population densities (e.g.,bioengineering techniques developed for cellcultures). A stream ecologist wishing to applythese techniques faces problems of interferencefrom dissolved and particulate substances andof low numbers of organisms. This is just oneof many instances where a cross-disciplinarydiscussion of methods would seem a useful firststep. Some of the problems discussed in thispaper that would benefit from a workshop onmethods include biologically meaningful frac-tionation of organics, spiraling measures, andassessment of elemental dynamics within bio-films and exchanges between biofilms and thewater column. As an example, a workshop ad-dressing the question "How do we assess nu-trient availability in biofilms in streams?" wouldbe of interest because biofilms appear to be oneof the critical interfaces controlling elementaldynamics in streams. Disciplines that should berepresented at such a workshop include: ana-lytical chemistry, hydraulic engineering,biomedical engineering, biophysics, chemicalengineering, and image analysis.
Biological centers program
At least four possibilities for stream ecologycenters are worthy of further discussion amongstream ecologists. The centers we propose arefor the most part facilities for field research.Although these centers may be quite differentfrom those conceived in other disciplines (e.g.,centers built around a supercomputer or so-phisticated network of telescopes), it is appro-priate in a science dedicated to understandingthe natural environment.
Any of the centers would be open to the en-tire community of stream researchers. A usefulanalogy might be the fleet of ships maintained
by the National Science Foundation and avail-able for oceanographic research. Just as anoceanographer requests ship-time, a streamecologist could request "stream-time." Al-though a core staff would be necessary to main-tain the monitoring base or the major long-termmanipulations at the centers (like the crew ona ship), the management of these facilitiesshould maximize the research done by scientistswho are not employees of the center.
A key to the success of such a venture wouldbe the position of research coordinator. Thisperson will need to coordinate work of indi-vidual investigators and promote cooperativeresearch between individuals working on dif-ferent aspects of stream ecology. If full use isto be made of the available experimental treat-ments, the coordinator should insist that in-vestigators put their major effort into definingthe impact of the experimental treatment on theaspect of the system that interests them, be itnutrient uptake kinetics, fish species diversity,or something else. Investigators wouldn't nec-essarily have to work together as long as theyfocus on the experiment and work at the rightspatial and temporal scale. Investigators wouldneed to communicate their results quickly tothe coordinator to facilitate synthesis and plan-ning of future experiments.
We propose here four potential stream ecol-ogy centers, listed in order of increasing poten-tial contribution to the field:
(1) Artificial Channel Reality Check (ACRC).—There are problems for which manipulation ofnatural streams will be either unfeasible orunethical (e.g., toxicant release). Artificialstreams offer an alternative approach to exam-ining lotic ecosystems. Artificial stream studieshave made significant contributions to the fieldof stream ecology (e.g., Mclntire 1973), but theircosts are such that individual investigators orinstitutions may not be able to afford ones thattruly mimic a natural channel. An artificialstreams center could provide such a facility. Onegoal of this center would be to develop artificialchannels that can be controlled and manipu-lated and that mimic natural channels in char-acteristics such as hydraulics, geomorphology,water quality, temperature, light quantity andquality, and characteristics of the biologicalcommunity. Hence a "reality check" would beone mission of this type of center: How do wedesign artificial channels that truly mimic nat-
1988] ELEMENTAL DYNAMICS IN STREAMS 425
ural streams? To what extent is exact duplica-tion of the taxonomic composition of naturalchannels necessary? Other relevant questionsinclude: At what scale are artificial streams ap-propriate? For example, they could be quite use-ful for examining phenomena in biofilms, butnot useful for examining watershed-level phe-nomena. When evaluating one parameter, areother factors kept constant or allowed to vary?An artificial stream center might also be an ap-propriate place to do pilot studies on methodsdevelopment. The authors do not agree on theusefulness of artificial channels. Some considerthem a distraction that siphons off resourcesand talent; others find them to be one of severaluseful approaches to studying streams.
(2) Individual Collaboration on Streams and Rivers(1CSR).— This alternative could be thought ofas enrichment of the status quo. Rather thandeveloping new stream research centers,another alternative would be to encourage newgrant applications for collaborative research onstreams. Teams of researchers could tackle aproblem on a diversity of levels encompassingphysiological, population, community, and eco-system ecology. If expertise from each of thesesubdisciplines were directed at a common wholestream experiment (e.g., controls on and effectsof nutrient supply and availability in a stream),leaps in understanding should be possible. TheRiver Continuum Study is an example of howthis type of cooperative study can make signif-icant contributions to the field, although not allsubdisciplines were involved in this project andit was descriptive rather than experimental.
(3) Experimental Streams and Rivers Area(ESRA).—The Experimental Lakes Area in On-tario has made a significant contribution to len-tic ecology through its effective use of wholeecosystem experiments. A center that has underits control several manipulatable lotic ecosys-tems, and a strong research coordinator who isable to attract wide participation in the exper-iments from scientists around the world, couldmake significant contributions to the field ofstream ecology. The center should have a strongsupport staff and a budget to maintain the fa-cility and keep the experiments running. Asdiscussed above, the scientific research shouldnot be done by center staff, but rather by sci-entists from .the community at large. The centershould have at its disposal the funds to supportsalaries of visiting investigators for periods up
to a year. This would offer scientists from bothresearch and teaching institutions a place andfinancial support to study the effect of the ex-perimental manipulation (e.g., nutrient enrich-ment or predator removal) on the processes ororganisms that interest them. There are severalkeys to the success of a project like this: theresearch coordinator, accessibility to the sci-entific community, adequate maintenance andrepair program, and a strong technical supportstaff. A major drawback to this kind of a centeris that resources would be concentrated at a sitein a single geographic region and hence con-clusions might not be applicable in a wide rangeof streams.
(4) Benchmark Streams and Rivers (BSR).—Analternative to developing a single experimentalstreams area would be to develop a network ofstreams and rivers in a range of latitudes andhydrologic regimes that would be amenable tomanipulation. If we were able to do wholestream nutrient enrichment experiments in arepresentative range of lotic ecosystems wewould make significant progress in understand-ing the scientific questions posed in this paper.The needs for a series of benchmark streamswould be similar to those described above: astrong research coordinator, accessibility to thescientific community, adequate maintenanceand repair program, and a strong technical sup-port staff. The U.S. Geological Survey has es-tablished a Hydrologic Benchmark Network thatincludes 57 streams across the U.S. in differentphysiographic regions with relatively protectedwatersheds and different hydrologic (and hencedisturbance) regimes (Cobb and Biesecker 1971;M. E. Gurtz, U.S. Geological Survey, personalcommunication; Resh et al. 1988—see this is-sue). These offer a wide range of streams tochoose from in the U.S., although it would benecessary to extend this network to arctic andtropical regions not currently part of the Hy-drologic Benchmark Network. Several excellentresearch stations that already exist in these re-gions would be appropriate for the internation-al cooperative research envisioned here. Firstthrough fifth- or sixth-order streams should beincluded in this network, i.e., work should notbe done exclusively in first- and second-orderstreams. The major focus of a network of ex-perimental stream centers would be on com-parative stream ecology over gradients of lati-tude, stream size, and disturbance regime. The
426 J. L. MEYER ET AL. [Volume 7
research coordinators would need to ensure thatthe experiments were of the appropriate natureand scale to provide comparable results.
Summary: future research directions
We have argued that controls on elementalsupply and availability in streams include bothlandscape-level and within-stream processesthat operate on a range of temporal and spatialscales. The temporal scales about which we knowthe least lie at both extremes: long-term (decadeto century) processes and short-term events suchas storms. One fruitful direction for further re-search is examining the impact of short-termevents like storms on element availability to thebiological community, and how that impactcompares with other physical and biologicalprocesses regulating elemental dynamics. A keyquestion in future studies of elemental dynam-ics in streams at all scales is the relative avail-ability of the various forms of an element to thebiological community. We need to move be-yond combining many forms of an element intoa single operationally-defined fraction (e.g.,DOC), which is a composite of forms varyinggreatly in their availability to the biota; insteadwe should concentrate on individual com-pounds or groups of compounds.
It is critical that future stream research ad-dress the question of the relative importance oflongitudinal linkages (upstream-downstream),lateral linkages (riparian, floodplain), and with-in-stream recycling as regulators of elementsupply to the biota. To what extent and on whattemporal and spatial scales is the biological re-quirement for various elements met by internalrecycling vs. lateral or longitudinal sources?
Our understanding of the controls on ele-mental supply and availability will be en-hanced by studies in several key areas. The con-cept of nutrient spiraling has facilitated studyof elemental dynamics in streams and couldfruitfully be applied to a greater variety ofstreams to further our understanding of link-ages between structure and function in streamcommunities. The extent to which the microbialfood loop is an important regenerator of nu-trients is unknown for stream ecosystems. Sev-eral poorly understood ecotones in streams aresites in which critical processes occur that maycontrol elemental supply and availability to thebiota. These include biofilms, the hyporheic
zone, floodplains, and zones of transition be-tween anaerobic and aerobic conditions. Theextent to which elemental bioavailability instreams is influenced by interactions among ele-ments is also not well understood.
Availability of elements has been demon-strated to regulate decomposition and primaryproductivity in some streams. The impact of thison higher trophic levels is less clearly under-stood. A better understanding of the interac-tions between elements and key ecological pro-cesses in streams could be achieved by the useof whole system manipulations supplementedwith intensive studies of key processes and pa-rameters. A network of experimental streamsover a wide range of latitudes and disturbanceregimes would allow researchers to addressthese critical questions in stream ecology.
National Science Foundation grant BSR-8611843 provided the financial support for theworkshop in which these ideas were developedand for the publication costs. The followingpeople made useful suggestions during discus-sion of these topics: G. Bretschko, W. J. Mat-thews, J. R. Webster, K. Suberkropp, P. B. Moyle,D. L. Correll, and T. L. Dudley. The manuscriptbenefitted from comments of A. P. Covich andthe Meyerfauna.
ALLAN, J. D. 1983. Predator-prey relationships instreams. Pages 191-230 in ]. R. Barnes and G. W.Minshall (editors). Stream ecology: applicationand testing of general ecological theory. PlenumPress, New York.
BALL, R. C, AND T. G. BAHR. 1975. Intensive survey:Red Cedar River, Michigan. Pages 431-460 in B.A. Whitton (editor). River ecology. University ofCalifornia Press, Berkeley.
BARSDATE, R. J., R. T. PRENTKI, AND T. FENCHEL. 1974.Phosphorus cycle of model ecosystems: signifi-cance for decomposer food chains and effects ofbacterial grazers. Oikos 25:238-251.
BARTLETT, K. B., P. M. CRILL, D. I. SEBACHER, R. C.HARRISS, J. O. WILSON, AND J. M. MELACK. 1988.Methane flux from the Central Amazonian flood-plain. Journal of Geophysical Research 93:1571-1582.
BILBY, R. E., AND G. E. LIKENS. 1980. Importance oforganic debris dams in the structure and functionof stream ecosystems. Ecology 61:1107-1113.
1988] ELEMENTAL DYNAMICS IN STREAMS 427
BOTT, T. L. 1983. Primary productivity in streams.Pages 29-54 in J. R. Barnes and G. W. Minshall(editors). Stream ecology: application and testingof general ecological theory. Plenum Press, NewYork.
BOTT, T. L., L. A. KAPLAN, AND F. T. KUSERK. 1984.Benthic bacterial biomass supported by stream-water dissolved organic matter. Microbial Ecol-ogy 10:335-344.
BRETSCHKO, G., AND M. LHCHTFRIED. 1988. Distri-bution of organic matter and fauna in a secondorder alpine gravel stream (Ritrodat-Lunz StudyArea, Austria). Verhandlungen der Internatio-nalen Vereinigung fur Theoretische und Ange-wandte Limnologie 23 (in press).
CARPENTER, S. R., J. F. KTTCHELL, AND J. R. HODGSON.1985. Cascading trophic interactions and lakeproductivity. BioScience 35:634-639.
COBB, E. D., AND J. E. BIESECKER. 1971. The NationalHydrologic Bench-Mark Network. GeologicalSurvey Circular 460-D. U.S. Geological Survey,Washington, D.C.
CORRELL, D. L., N M. GOFF, AND W. T. PETERJOHN.1984. Ion balances between precipitation inputsand Rhode River watershed discharges. Chapter5 in O. P. Bricker (editor). Geological aspects ofacid deposition. Butterworth Publishers, Boston.
CORRELL, D. L., J. J. MIKLAS, A. H. MINES, AND J. J.SCHAFER. 1987. Chemical and biological trendsassociated with acidic atmospheric deposition inthe Rhode River watershed and estuary. Water,Air and Soil Pollution 35:63-86.
COSBY, B. J., G. M. HORNBERGER, J. GALLOWAY, AND R.F.WRIGHT. 1985. Time scales of catchment acid-ification. Environmental Science and Technology19:1144-1149.
COVICH, A. 1988. Geographical and historical com-parisons of neotropical streams: biotic diversityand detrital processing in highly variable habi-tats. Journal of the North American Bentholog-ical Society 7:361-386.
CROCKER, M. T., AND J. L. MEYER. 1987. Interstitialdissolved organic carbon in sediments of a south-ern Appalachian headwater stream. Journal ofthe North American Benthological Society 6:159-167.
DAHM, C. N., E. H. TROTTER, AND J. R. SEDELL. 1987.Role of anaerobic zones and processes in streamecosystem productivity. Pages 157-178 in R. C.Averett and D. M. McKnight (editors). Chemicalquality of water and the hydrologic cycle. LewisPublishers, Chelsea, Michigan.
DELUCCHI, C. M. 1988. Comparison of communitystructure among streams with different temporalflow regimes. Canadian Journal of Zoology 66:579-586.
DILLON, P. J., AND W. B. KIRCHNER. 1975. The effectsof geology and land use on the export of phos-
phorus from watersheds. Water Research 9:135-148.
DUCKLOW, H. W., D. A. PURDIE, P. J. LE B. WILLIAMS,ANDj.M. DAVIES. 1986. Bacterioplankton: a sinkfor carbon in a coastal marine plankton com-munity. Science 232:865-867.
EDWARDS, R. T. 1987. Seasonal bacterial biomass dy-namics in the seston of two southeastern black-water rivers. Limnology and Oceanography 32:221-234.
EDWARDS, R. T., AND J. L. MEYER. 1986. Productionand turnover of planktonic bacteria in two south-eastern blackwater rivers. Applied and Environ-mental Microbiology 52:1317-1323.
EDWARDS, R. T., AND J. L. MEYER. 1987a. Metabolismof a sub-tropical low-gradient blackwater river.Freshwater Biology 17:251-263.
EDWARDS, R. T., AND J. L. MEYER. 1987b. Bacteria asa food source for blackfly larvae in a blackwaterriver. Journal of the North American Bentho-logical Society 6:241-250.
ELWOOD, J. W., P. J. MULHOLLAND, AND J. D. NEWBOLD.1988. Microbial activity and P uptake on decom-posing leaf detritus in a heterotrophic stream.Verhandlungen der Internationalen Vereini-gung fur Theoretische und Angewandte Lim-nologie 23 (in press).
ELWOOD, J. W., J. D. NEWBOLD, R. V. O'NEILL, AND W.VAN WINKLE. 1983. Resource spiraling: an op-erational paradigm for analysing lotic ecosys-tems. Pages 3-28 in T. D. Fontaine and S. M. Bar-tell (editors). Dynamics of lotic ecosystems. AnnArbor Science, Ann Arbor, Michigan.
ELWOOD, J. W., J. D. NEWBOLD, A. F. TRIMBLE, AND R.W. STARK. 1981. The limiting role of phos-phorus in a woodland stream ecosystem: effectsof P enrichment on leaf decomposition and pri-mary producers. Ecology 62:146-158.
ELWOOD, J. W., AND R. R. TURNER. 1988. Streams:water chemistry and ecology. In D. W. Johnsonand R. I. Van Hook (editors). Biogeochemical cy-cling in Walker Branch watershed: a synthesis ofresearch results. Springer-Verlag, New York (inpress).
EPPLEY, R. W., AND B. J. PETERSON. 1979. Participateorganic matter flux and planktonic new produc-tion in the deep ocean. Nature 282:677-680.
FENCHEL, T. 1970. Studies on the decomposition oforganic detritus derived from the turtle grass,Thalassia testudinum. Limnology and Oceanogra-phy 15:14-20.
FENCHEL, T. 1978. The ecology of micro- and meio-benthos. Annual Review of Ecology and System-atics 9:99-121.
FINDLAY, S., J. L. MEYER, AND P. J. SMITH. 1986. In-corporation of microbial biomass by Peltoperla sp.(Plecoptera) and Tipula sp. (Diptera). Journal of
428 J. L. MEYER ET AL. [Volume 7
the North American Benthological Society 5:306-310.
FISHER, S. G., L. J. GRAY, N. B. GRIMM, AND D. E. BUSCH.1982. Temporal succession in a desert streamecosystem following flash flooding. EcologicalMonographs 52:93-110.
FISHER, S. G., AND G. E. LIKENS. 1973. Energy flowin Bear Brook, New Hampshire: an integrativeapproach to stream ecosystem metabolism. Eco-logical Monographs 43:421-439.
FURCH, K. 1984. Water chemistry of the AmazonBasin: the distribution of chemical elementsamong freshwaters. Pages 196-200 in H. Sioli (ed-itor). The Amazon: limnology and landscapeecology of a mighty tropical river and its basin.Dr. W. Junk Publishers, Dordrecht.
GAUDET, J. J., AND J. M. MELACK. 1982. Major ionchemistry in a tropical African lake basin. Fresh-water Biology 11:309-333.
GOLDMAN, J. C, AND D. A. CARON. 1985. Experi-mental studies on an omnivorous microflagellate:implications for grazing and nutrient regenera-tion in the marine microbial food chain. DeepSea Research 32:899-915.
GREGORY, S. V. 1983. Plant-herbivore interactionsin stream systems. Pages 159-189 in J. R. Barnesand G. W. Minshall (editors). Stream ecology: ap-plication and testing of general ecological theory.Plenum Press, New York.
GRIMM, N. B., AND S. G. FISHER. 1984. Exchange be-tween interstitial and surface water: implicationsfor stream metabolism and nutrient cycling. Hy-drobiologia 111:219-228.
GRIMM, N. B., AND S. G. FISHER. 1986. Nitrogen lim-itation in a Sonoran desert stream. Journal of theNorth American Benthological Society 5:2-15.
GURTZ, M. E., J. R. WEBSTER, AND J. B. WALLACE. 1980.Seston dynamics in southern Appalachianstreams: effects of clear-cutting. Canadian Jour-nal of Fisheries and Aquatic Sciences 37:624-631.
HART, D. D. 1981. Foraging and resource patchiness:field experiments with a grazing stream insect.Oikos 37:46-52.
HENDERSON, G. S., AND W. F. HARRIS. 1975. An eco-system approach to characterization of the nitro-gen cycle in a deciduous forest watershed. Pages179-193 in B. Bernier and C. H. Winget (editors).Forest soils and forest land management. LavalUniversity Press, Quebec.
HILL, A. R. 1979. Denitrification in the nitrogen bud-get of a river ecosystem. Nature 281:291-292.
HILL, A. R. 1983. Nitrate-nitrogen mass balances fortwo Ontario rivers. Pages 457-477 in T. D. Fon-taine and S. M. Bartell (editors). Dynamics of loticecosystems. Ann Arbor Science Publishers, AnnArbor, Michigan.
HORNE, A. J. 1975. Algal nitrogen fixation in Cali-
fornia streams: diel cycles and nocturnal fixation.Freshwater Biology 5:471-477.
HORNE, A. J., AND C. J. W. CARMIGGELT. 1975. Algalnitrogen fixation in California streams: seasonalcycles. Freshwater Biology 5:461-470.
HYNES, H. B. N. 1969. The enrichment of streams.Pages 188-196 in Eutrophication: causes, conse-quences, corrections. National Academy of Sci-ences, Washington, D.C.
HYNES, H. B. N. 1983. Groundwater and stream ecol-ogy. Hydrobiologia 100:93-99.
JOHANNES, R. E. 1965. Influence of marine protozoaon nutrient regeneration. Limnology and Ocean-ography 10:434-442.
JOHNSON, N. M., C. T. DRISCOLL, J. S. EATON, G. E.LIKENS, AND W. H. MCDOWELL. 1981. "Acid rain,"dissolved aluminum and chemical weathering atthe Hubbard Brook Experimental Forest, NewHampshire. Geochemica et Cosmochimica Acta45:1421-1437.
JOHNSON, N. M., G. E. LIKENS, F. H. BORMANN, D. W.FISHER, AND R. S. PIERCE. 1969. A working modelfor the variation in stream water chemistry at theHubbard Brook Experimental Forest, NewHampshire. Water Resources Research 5:1353-1363.
JORDAN, C. F. (editor). 1987. Amazonian rain forests:ecosystem disturbance and recovery. EcologicalStudies 60. Springer-Verlag, New York.
JORGENSEN, B. B.,ANDN. P.REVSBECH. 1985. Diffusiveboundary layers and the oxygen uptake of sed-iments and detritus. Limnology and Oceanog-raphy 30:111-122.
JUNK, W. J. 1985. The Amazon floodplain—a sink orsource for organic carbon. Mitteilungen aus demGeologische-Palaontologischen Institut der Uni-versitat Hamburg. SCOPE/UNEP Sonderband 58:267-283.
KAPLAN, L. A. 1980. Dissolved organic matter in arural piedmont watershed. Ph.D. Thesis, Uni-versity of Pennsylvania, Philadelphia.
KAPLAN, L. A., AND T.L.BoTT. 1982. Diel fluctuationsof DOC generated by algae in a piedmont stream.Limnology and Oceanography 27:1091-1100.
KAPLAN, L. A., R. A. LARSON, AND T. L. BOTT. 1980.Patterns of dissolved organic carbon in transport.Limnology and Oceanography 25:1034-1043.
KELLMAN, M., J. HUDSON, AND K. SANMUGADAS. 1982.Temporal variability in atmospheric nutrient in-flux to a tropical ecosystem. Biotropica 14:1-9.
KOHLER, S. L. 1984. Search mechanism of a streamgrazer in patchy environments: the role of foodabundance. Oecologia 62:209-218.
KRAMER, J. R., A. W. ANDREN, R. A. SMITH, A. H.JOHNSON, R. B. ALEXANDER, AND G. OEHLERT. 1986.Streams and lakes. Pages 231-299 in Acid depo-sition—long-term trends. National Academy of
1988] ELEMENTAL DYNAMICS IN STREAMS 429
Sciences, National Academy Press, Washington,D.C.
LAY, J. A., AND A. K. WARD. 1987. Algal communitydynamics in two streams associated with differ-ent geological regions in the southeastern UnitedStates. Archiv fur Hydrobiologie 108:305-324.
LmcHTFWED, M. 1985. Organic matter in gravelstreams (Project Ritrodat-Lunz). Verhandlungender Internationalen Vereinigung fur Theore-tische und Angewandte Limnologie 22:2058-2062.
LESACK, L. F. W., R. E. HECKY, AND J. M. MELACK. 1984.Transport of carbon, nitrogen, phosphorus andmajor solutes in the Gambia River, West Africa.Limnology and Oceanography 29:810-830.
LEWIS, W. M. 1986. Nitrogen and phosphorus runofflosses from a nutrient-poor tropical moist forest.Ecology 67:1275-1828.
LEWIS, W. M., AND M. C. GRANT. 1979. Relationshipsbetween stream discharge and yield of dissolvedsubstances from a Colorado mountain watershed.Soil Science 128:353-363.
LEWIS, W. M., J. F. SAUNDERS, S. LEVINE, AND F. H.WEIBEZAHN. 1986. Organic carbon in the CauraRiver, Venezuela. Limnology and Oceanography31:653-656.
LIKENS, G. E., F. H. BORMANN, N. M. JOHNSON, D. W.FISHER, AND R. S. PIERCE. 1970. Effects of forestcutting and herbicide treatment on nutrient bud-gets in the Hubbard Brook Watershed ecosystem.Ecological Monographs 40:23-47.
LIKENS, G/E., F. H. BORMANN, N. M. JOHNSON, ANDR. S. PIERCE. 1967. The calcium, magnesium, po-tassium, and sodium budgets for a small forestedecosystem. Ecology 48:772-785.
LIKENS, G. E., F. H. BORMANN, R. S. PIERCE, AND J. S.EATON. 1985. The Hubbard Brook valley. Pages9-39 in G. E. Likens (editor). An ecosystem ap-proach to aquatic ecology. Springer-Verlag, NewYork.
LIKENS,^. E., F. H. BORMANN, R. S. PIERCE, J. S. EATON,AND N. M. JOHNSON. 1977. Biogeochemistry ofa forested ecosystem. Springer-Verlag, New York.
LOCK, M. A., R. R. WALLACE, J. W. COSTERTON, R. M.VENTULLO, AND S. E. CHARLTON. 1984. River epi-lithon: toward a structural-functional model. Oi-kos 42:10-22.
MASON, J., AND H. M. SHIP. 1985. The current stateof knowledge on acidification of surface watersand guidelines for further research. Ambio 14:45-51.
MCDOWELL, W. H. 1984. Water quality in rain foreststreams of Puerto Rico. Pages 94-107 in Proceed-ings of the First Caribbean Islands Water Re-sources Congress. Water Resources Research In-stitute, University of Puerto Rico, Mayaguez.
MCDOWELL, W. H., AND G. E. LIKENS. 1988. Origin,composition, and the flux of dissolved organic
carbon in the Hubbard Brook Valley. EcologicalMonographs 58:177-195.
MCDOWELL, W. H., AND T. WOOD. 1984. Podzoliza-tion: soil processes control dissolved organic car-bon concentrations in stream water. Soil Science137:23-32.
MclNTTRE, C. D. 1973. Periphyton dynamics in lab-oratory streams: a simulation model and its im-plications. Ecological Monographs 43:399-420.
MEYER, J.L. 1988. Benthic bacterial biomass and pro-duction in a blackwater river. Verhandlungender Internationalen Vereinigung fiir Theore-tische und Angewandte Limnologie 23 (in press).
MEYER, J. L., R. T. EDWARDS, AND R. RISLEY. 1987a.Bacterial growth on dissolved organic carbon froma blackwater river. Microbial Ecology 13:13-29.
MEYER, J. L., AND C. JOHNSON. 1983. The influenceof elevated nitrate concentration on leaf decom-position in a stream. Freshwater Biology 13:177-183.
MEYER, J. L., AND G. E. LIKENS. 1979. Transport andtransformation of phosphorus in a forest streamecosystem. Ecology 60:1255-1269.
MEYER, J. L., AND C. TATE. 1983. The effects of wa-tershed disturbance on dissolved organic carbondynamics of a stream. Ecology 64:33-44.
MEYER, J. L., C. M. TATE, R. T. EDWARDS, AND M.T. CROCKER. 1987b. The trophic significance ofDOC in streams. Pages 269-278 in W. T. Swankand D. A. Crossley (editors). Forest hydrologyand ecology at Coweeta. Springer-Verlag, NewYork.
MINSHALL, G. W., K. W. CUMMINS, R. C. PETERSON, C.E. GUSHING, D. A. BRUNS, J. R. SEDELL, AND R. L.VANNOTE. 1985. Developments in stream eco-system theory. Canadian Journal of Fisheries andAquatic Sciences 42:1045-1055.
MINSHALL, G. W., R. C. PETERSON, K. W. CUMMINS, T.L. BOTT, J. R. SEDELL, C. E. GUSHING, AND R. L.VANNOTE. 1983. Interbiome comparison ofstream ecosystem dynamics. Ecological Mono-graphs 53:1-25.
MITCH, W. J., AND J. GOSSEUNK. 1986. Wetlands. VanNostrand Reinhold Co., New York.
MORRISSEY K. M., AND T. R. FISHER. 1988. Regener-ation and uptake of ammonium by plankton inan Amazon floodplain lake. Journal of PlanktonResearch 10:31-48.
MOSES, B. S. 1987. The influence of flood regime onfish catch and fish communities of the Cross Riverfloodplain ecosystem, Nigeria. EnvironmentalBiology of Fishes 18:51-65.
MULHOLLAND, P. J., J. W. ELWOOD, J. D. NEWBOLD, ANDL. A. FERREN. 1985a. Effect of a leaf shreddinginvertebrate on organic matter dynamics andphosphorus spiraling in heterotrophic laboratorystreams. Oecologia 66:199-206.
430 J. L. MEYER ET AL. [Volume 7
MULHOLLAND, P. J., R. A. MlNEAR, AND J. W. ElAVOOD.1988. Production of soluble, high molecularweight phosphorus and its subsequent uptake bystream detritus. Verhandlungen der Internatio-nalen Vereinigung fur Theoretische und Ange-wandte Limnologie 23 (in press).
MULHOLLAND, P. J., J. D. NEWBOLD, J. W. ELWOOD, ANDC. L. HOM. 1983. The effect of glazing intensityon phosphorus spiralling in autotrophic streams.Oecologia 53:358-366.
MULHOLLAND, P. J., J. D. NEWBOLD, J. W. ELWOOD, ANDJ. R. WEBSTER. 1985b. Phosphorus spiralling ina woodland stream: seasonal variations. Ecology66:1012-1023.
NAIMAN, R. J., H. DECAMPS, J. PASTOR, AND C. A.JOHNSON. 1988. The potential importance ofboundaries to fluvial ecosystems. Journal of theNorth American Benthological Society 7:289-306.
NAIMAN, R. J., J. M. MELILLO, AND J. E. HOBBIE. 1986.Ecosystem alteration of boreal forest streams bybeaver (Castor ccmadensis). Ecology 67:1254-1269.
NEILL, W. E. 1981. Impact of Chaoborus predationupon the structure and dynamics of a crustaceanzooplankton community. Oecologia 48:164-177.
NEILL, W. E., AND A. PEACOCK. 1979. Breaking thebottleneck: interactions of invertebrate predatorsand nutrients in oligotrophic lakes. Pages 715-724 in W. C. Kerfoot (editor). Evolution and ecol-ogy of zooplankton communities. University ofNew England Press, Hanover, New Hampshire.
NEWBOLD, J. D., J. W. ELWOOD, R. V. O'NEILL, AND A.L. SHELDON. 1983. Phosphorus dynamics in awoodland stream ecosystem: a study of nutrientspiralling. Ecology 64:1249-1265.
NEWBOLD, J. D., J. W. ELWOOD, R. V. O'NEILL, AND W.VAN WINKLE. 1981. Measuring nutrient spiral-ling in streams. Canadian Journal of Fisheriesand Aquatic Sciences 38:860-863.
NEWBOLD, J. D,, P. J. MULHOLLAND, J. W. ELWOOD, ANDR. V. O'NEILL. 1982a. Organic carbon spirallingin stream ecosystems. Oikos 38:266-272.
NEWBOLD, J. D., R. V. O'NEILL, J. W. ELWOOD, AND W.VAN WINKLE. 1982b. Nutrient spiralling in.streams: implications for nutrient limitation andinvertebrate activity. American Naturalist 120:628-652.
OBERNDORFER, R., J. V. MCARTHUR, J. R. BARNES, ANDJ. DIXON. 1984. The effect of invertebrate pred-ators on leaf litter processing in an alpine stream.Ecology 65:1325-1331.
O'DOHERTY, E. C. 1985. Stream-dwelling copepods:their life history and ecological significance. Lim-nology and Oceanography 30:554-564.
OMERNIK, J. M. 1977. Nonpoint source-stream nu-trient level relationships: a nationwide survey.U.S. Environmental Protection Agency, Corval-lis, Oregon, EPA-600/3-77-105.
PALUMBO, A. V., M. A. BOGLE, R. R. TURNER, J. W.
ELWOOD, AND P. J. MULHOLLAND. 1987. Bacterialcommunities in acidic and circumneutral streams.Applied and Environmental Microbiology 53:337-344.
PAOUNI, J., R. HERRERA, AND A. NEMETH. 1983. Hy-drochemistry of the Orinoco and Caroni River.Mitteilungen aus dem Geologisch-Paleontolo-gischen Institut der Universitat Hamburg.SCOPE/UNEP Sonderband 55:223-236.
PATRICK, R. 1970. Benthic stream communities.American Scientist 58:546-549.
PATRICK, R. 1978. Effects of trace metals in the aquat-ic ecosystem. American Scientist 66:185-191.
PECKARSKY, B. L. 1980. Predator-prey interactionsbetween stoneflies and mayflies: behavioral ob-servations. Ecology 61:932-943.
PECKARSKY, B. L. 1982. Aquatic insect predator-preyrelations. BioScience 32:261-266.
PECKARSKY, B. L. 1984. Predator-prey interactionsamong aquatic insects. Pages 196-254 in V. Reshand D. Rosenberg (editors). The ecology of aquat-ic insects. Praeger Publishers, New York.
PETERJOHN, W. T., AND D. C. CORRELL. 1984. Nutrientdynamics in an agricultural watershed: obser-vations on the role of a riparian forest. Ecology65:1466-1475.
PETERSON, B. J., J. E. HOBBIE, T. L. CORLISS, AND K.KRIET. 1983. A continuous-flow periphytonbioassay: tests of nutrient limitation in a tundrastream. Limnology and Oceanography 28:583-591.
PETERSON, B. J., J. E. HOBBIE, A. E. HERSHEY, M. A.LOCK, T. E. FORD, J. R. VESTAL, V. L. MCKINLEY,M. A. J. HULLAN, M. C. MILLER, R. M. VENTULLO,AND G. S. VOLK. 1985. Transformation of a tun-dra river from heterotrophy to autotrophy by ad-dition of phosphorus. Science 229:1383-1386.
POMEROY, L. R. 1974. The ocean's food web, a chang-ing paradigm. BioScience 24:499-504.
POWER, M. E., W. J. MATTHEWS, AND A. J. STEWART.1985. Grazing minnows, piscivorous bass, andstream algae: dynamics of a strong interaction.Ecology 66:1448-1456.
PRINGLE, C. M. 1985. Effects of chironomid (Insecta:Diptera) tube-building activities on stream dia-tom communities. Journal of Phycology 21:185-194.
PRINGLE, C. M. 1987. Effects of water and substratumnutrient supplies on lotic periphyton growth: anintegrated bioassay. Canadian Journal of Fish-eries and Aquatic Sciences 44:619-629.
PRINGLE, C. M., AND J. A. BOWERS. 1984. An in situsubstratum fertilization technique: diatom colo-nization on nutrient-enriched, sand substrate.Canadian Journal of Fisheries and Aquatic Sci-ences 41:1247-1251.
PRINGLE, C. M., R. J. NAIMAN, G. BRETSCHKO, J. R. KARR,M. W. OSWOOD, J. R. WEBSTER, R. L. WELCOMME,AND M. J. WINTERBOURN. 1988. Patch dynamics
1988] ELEMENTAL DYNAMICS IN STREAMS 431
in lotic systems: the stream as a mosaic. Journalof the North American Benthological Society 7:503-524.
PRINGLE, C. M., P. PAABY-HANSEN, P. D. VAUX, AND C.R. GOLDMAN. 1986. In situ nutrient assays ofperiphyton growth in a lowland Costa Ricanstream. Hydrobiologia 134:207-213.
RESH, V. H., A. V. BROWN, A. P. COVICH, M. E. GURTZ,H. W. Li, G. W. MINSHALL, S. R. RHCE, A. L.SHELDON, J. B. WALLACE, AND R. WISSMAR. 1988.The role of disturbance theory in stream ecology.Journal of the North American Benthological So-ciety 7:433-455.
RIGLER, F. H. 1979. The export of phosphorus fromDartmoor catchments: a model to explain varia-tions of phosphorus concentrations in streamswater. Journal of the Marine Biological Associ-ation of the United Kingdom 59:659-687.
ROUNICK, J. S., M. J. WlNTERBOURN, AND G. L. LYON.1982. Differential utilization of allochthonousand autochthonous inputs by aquatic inverte-brates in some New Zealand streams: a stablecarbon isotope study. Oikos 39:191-198.
RUBLEE, P. A., S. M. MERKEL, AND M. A. FAUST. 1983.Transport of bacteria in the sediments of a tem-perate marsh. Estuarine and Coastal Shelf Science16:501-509.
RUBLEE, P. A., S. M. MERKEL, M. A. FAUST, AND J. MIK-LAS. 1984. Distribution and activity of bacteriain the headwaters of the Rhode River estuary,Maryland, USA. Microbial Ecology 10:243-255.
RUTHERFORD, J. E., AND H. B. N. HYNES. 1987. Dis-solved organic carbon in streams and ground-water. Hydrobiologia 154:33-48.
SANHUEZA, E., G. CUENCA, M. J. GOMEZ, R. HERRERA,C. ISHIZAKI, I. MARTI, ANDJ. PAOLINI. 1988. Char-acterization of the Venezuelan environment andits potential for acidification. Pages 197-255 in H.Rodhe and R. Herrera (editors). Acidification intropical countries. SCOPE 36. John Wiley andSons Ltd., New York.
SCHLESINGER, W. H., AND J. M. MELACK. 1981. Trans-port of organic carbon in the world's rivers. Tel-lus 33:172-187.
SEDELL, J. R., AND J. E. RICKEY. 1988. The river con-tinuum concept: a basis for the expected ecosys-tem behavior of very large rivers. Canadian Jour-nal of Fisheries and Aquatic Sciences (in press).
SEGARS, J. E., R. A. MINEAR, J. W. ELWOOD, AND P. J.MULHOLLAND. 1986. Chemical characterizationof soluble phosphorus forms along a hydrologicflow path of a forested stream ecosystem. OakRidge National Laboratory, ORNL/TM-9737.
SHERR, E. B., B. F. SHERR, AND L. ALBRIGHT. 1987.Bacteria: link or sink. Science 235:88-89.
SIOLI, H. 1975. Tropical rivers as expressions of theirterrestrial environments. Pages 275-288 in F. B.Golley and E. Medina (editors). Ecological studies
Siou, H. (editor). 1984. The Amazon: limnology andlandscape ecology of a mighty tropical river andits basin. Dr. W. Junk Publishers, Dordrecht.
STALLARD, R. F. 1985. River chemistry, geology, geo-morphology, and soils in the Amazon and Ori-noco Basins. Pages 293-316 in J. I. Drever (editor).The chemistry of weathering. Kluwer AcademicPublishers, Hingham, Massachusetts.
STALLARD, R. F., AND J. M. EDMOND. 1983. Geochem-istry of the Amazon. 2. The influence of geologyand weathering environment on the dissolvedload. Journal of Geophysical Research 88:9671.
SUBERKROPP, K. 1984. Effect of temperature on theseasonal occurrence of aquatic hypomycetes.Transactions of the British Mycological Society82:53-62.
SWANK, W. T., AND W. H. CASKEY. 1982. Nitrate de-pletion in a second-order mountain stream. Jour-nal of Environmental Quality 11:581-584.
SWANK, W. T., AND J. E. DOUGLASS. 1975. Nutrientflux in undisturbed and manipulated forest eco-systems in the southern Appalachian mountains.Publications de 1'Association Internationale desSciences Hydrologiques Symposium de Tokyo117:445-456.
THURMAN, E. M. 1985. Organic geochemistry of nat-ural waters. Martinus Nijhoff/Dr. W. Junk, Bos-ton.
TODD, R., R. LOWRANCE, O. HENDERICKSON, L.ASMUSSEN, R. LEONARD, J. FAIL, AND B. HERRICK.1983. Riparian vegetation as filters of nutrientsexported from a Coastal Plain agricultural wa-tershed. Pages 485-493 in R. Lowrance. R. Todd,L. Asmussen, and R. Leonard (editors). Nutrientcycling in agricultural ecosystems. University ofGeorgia College of Agriculture Experiment Sta-tion Special Publication. 23. Athens.
TRISKA, F. J., AND R. J. OREMLAND. 1981. Denitrifi-cation associated with periphyton communities.Applied and Environmental Microbiology 42:745-748.
TRISKA, F. J., J. R. SEDELL, K. CROMACK, S. V. GREGORY,AND F. M. McCoRlsON. 1984. Nitrogen budgetfor a small coniferous forest stream. EcologicalMonographs 54:119-140.
VANNOTE, R. L., G. W. MINSHALL, K. W. CUMMINS, J.R. SEDELL, AND C. E. GUSHING. 1980. The rivercontinuum concept. Canadian Journal of Fish-eries and Aquatic Science 37:130-137.
VENTULLO, R. M., AND J. J. ROWE. 1982. Denitrifica-tion potential of epilithic communities in a loticenvironment. Current Microbiology 7:29-34.
WALLACE, J. B., T. F. CUFFNEY, C. C. LAY, AND D. VOGEL.1987. The influence of an ecosystem-level ma-nipulation on prey consumption by a lotic dra-gonfly. Canadian Journal of Zoology 65:35-40.
432 J. L. MEYER ET AL. [Volume 7
WALLACE, J. B., J. R. WEBSTER, AND T. F. CUFFNEY. 1982.Stream detritus dynamics: regulation by inver-tebrate consumers. Oecologia 53:197-200.
WALLIS, P. M., H. B. N. HYNES, AND S. A. TELANG.1981. The importance of ground-water in thetransportation of allochthonous organic matterto the streams draining a small mountain basin.Hydrobiologia 79:77-90.
WEBSTER, J. R., AND E. F. BENFIELD. 1986. Vascularplant breakdown in freshwater ecosystems. An-nual Review of Ecology and Systematics 17:567-594.
WEBSTER, J. R., M. E. GURTZ, J. J. HAINS, }. L. MEYER,W. T. SWANK, J. B. WAIDE, AND J. B. WALLACE.1983. Stability of stream ecosystems. Pages 355-395 in J. R. Barnes and G. W. Minshall (editors).Stream ecology: application and testing of gen-eral ecological theory. Plenum Press, New York.
WELLER, D. E., W. T. PETERJOHN, N. M. GOFF, AND D.L. CORRELL. 1986. Ion and acid budgets for aforested Atlantic coastal plain watershed and theirimplications for the impact of acid deposition.Pages 392-421 in D. L. Correll (editor). Watershedresearch perspectives. Smithsonian InstitutionPress, Washington, D.C.
WILLIAMS, D. D. 1986. Factors influencing the mi-crodistributions of two sympatric species of Ple-coptera: an experimental study. Canadian Journalof Fisheries and Aquatic Sciences 43:1005-1009.
WOTTON, R. S. 1980. Bacteria as food for blackflylarvae (Diptera: Simuliidae) in a lake-outlet inFinland. Annales Zoologici Fennici 17:127-130.
WUHRMANN, K. 1972. Stream purification. Pages 119-151 in R. Mitchell (editor). Water pollution mi-crobiology. Wiley-Interscience, New York.