The Effects of System-Level Nutrient Enrichment on...

Journal of Coastal Research SI 45 110–133 West Palm Beach, Florida Fall 2004

The Effects of System-Level NutrientEnrichment on BacterioplanktonProduction in a Tidally-InfluencedEstuary*Jude K. Apple*†, P.A. del Giorgio‡, and R.I.E. Newell†

†Horn Point LaboratoryUniversity of Maryland

Center for EnvironmentalScience

2020 Horns Point RoadCambridge, MD 21613,U.S.A.

‡Departement des SciencesBiologiques

Universite du Quebec aMontreal

Case Postale 8888Succursale Centre-villeMontreal (Quebec),Canada H3C 3P8

ABSTRACT

APPLE, J.K.; DEL GIORGIO, P.A., and NEWELL, R.I.E., 2004. The Effects of System-Level NutrientEnrichment on Bacterioplankton Production in a Tidally Influenced Estuary. Journal of Coastal Research,SI(45), 110–133. West Palm Beach (Florida), ISSN 0749-0208.

We describe the use of Monie Bay as a natural experiment to evaluate the effect of system-level nutrientenrichment on natural bacterioplankton communities. Monie Bay, a component of the Chesapeake BayNational Estuarine Research Reserve System, is a subestuary of the Chesapeake Bay, consisting of a shal-low semienclosed bay and three tidally influenced creeks varying in their agricultural land use and fresh-water inputs. As part of a 2-year study in this system, we identified distinct spatial and seasonal patternsin ambient nutrient concentrations, salinity, and source and quantity of organic matter that were relatedto differences in agricultural practices and watershed characteristics among the three tidal creeks. Principalcomponents analysis identified freshwater delivery of nutrients and temperature as key factors driving theoverall variability of this system. Despite significant variability in nutrient concentrations and bacterio-plankton production (BP) throughout the year, we observed persistent response of bacterioplankton tonutrient enrichment, as evidenced by a comparison of 2-year averages in agriculturally developed LittleMonie Creek (LMC) relative to the undeveloped Little Creek (LC), and by a comparison of the nutrient-enriched upper estuary of LMC to sites nearer the open bay. Bacterioplankton responded positively topulsed nutrient availability, with elevated rates of BP associated with agriculturally derived nutrient inputsto the Monie Bay system. Freshwater inputs play an important role in mediating the response of bacter-ioplankton to nutrient enrichment, as evidenced by relatively low estimates of BP in the freshwater-dom-inated, agriculturally developed Monie Creek. This response is attributed to changes in organic matterquality in the system and the direct effect of salinity on bacterioplankton community metabolism.

ADDITIONAL INDEX WORDS: Bacterioplankton, bacterial production, nutrient loading, agricultural landuse, estuary, Monie Bay.

INTRODUCTION

Nonpoint source inputs of agriculturally derivednutrients have been unequivocally linked to nutri-ent enrichment and subsequent eutrophication ofcoastal systems (BEAULAC and RECKHOW, 1982;FISHER, 1985). Understanding the effect of nutri-

* Corresponding author. E-mail: [email protected]* This research was funded in part by a Graduate Re-

search Fellowship awarded by the National Estuarine Re-search Reserve System and by a grant from the Cooper-ative Institute for Coastal and Estuarine EnvironmentalTechnology, a partnership between the National Oceanicand Atmospheric Administration and the University ofNew Hampshire.

ent inputs on coastal system function is necessaryto allow agricultural production to be sustainedwithout sacrificing water quality and integrity ofestuarine resources. An integral program in thisendeavor is the National Estuarine Research Re-serve System (NERRS), a network of sites locatedthroughout the coastal US that have been estab-lished for long-term research, education, and stew-ardship of national estuarine resources. The use ofthese sites as ‘‘living laboratories’’ is a primary ob-jective of NERRS and plays an important role inunderstanding and mitigating anthropogenic im-pacts on the health and function of estuarine re-sources.

Journal of Coastal Research, Special Issue No. 45, 2004

111Bacterioplankton Production in a Tidally-Influenced Estuary

Coastal eutrophication is traditionally evaluatedwithin the context of phytoplankton abundance(SMITH, LEFFLER, and MACKIERNAN, 1992), in-creases in ambient nutrient concentrations and or-ganic matter loading (NIXON, 1995), and generalincreases in heterotrophic activity (TUTTLE, JO-NAS, and MALONE, 1987). Heterotrophic bacterio-plankton (i.e., the microbial community) are sel-dom incorporated in these assessments, despitetheir widespread acceptance as an important com-ponent in ecosystem health, function, and the eu-trophication process (DUCKLOW et al., 1986;SHERR and SHERR, 1988; BRUSSAARD and RIEG-MAN, 1998). The microbial community mediates al-most every important ecological process related toeutrophication of aquatic systems. Deep-water an-oxia and periods of net heterotrophy in the Ches-apeake Bay are driven almost exclusively by aer-obic, heterotrophic bacterial metabolism (TUTTLE,JONAS, and MALONE, 1987; SMITH and KEMP,2001). The transfer of algal-derived organic matterto top consumers and the extent to which it is lostvia respiratory processes are dictated by the effi-ciency of carbon cycling by the microbial commu-nity (AZAM et al., 1983; SHERR and SHERR, 1988).Similarly, the availability and cycling of inorganicnutrients are regulated by their rapid utilizationby bacterioplankton relative to phytoplankton(KIRCHMAN, 1994) and the variable efficiency withwhich dissolved organic matter (DOM) and partic-ulate organic matter (POM) are remineralized(DEL GIORGIO and COLE, 1998). The microbialcommunity not only drives these ecological pro-cesses, but also responds rapidly to even the mostsubtle changes in these processes (FINLAY, MA-BERLY, and COOPER, 1997). Thus, the character-istics of this community represent a sensitive andintegrative biological synthesis of environmentalconditions and ecological processes that effectivelyintegrates key ecological aspects of ecosystemfunction that are seldom considered in manage-ment and conservation efforts.

The response of natural bacterioplankton com-munities to inputs of inorganic nutrients has beenstudied extensively, although seldom on the spa-tial or temporal scale at which system-level nutri-ent enrichment typically occurs. Small-scale ex-periments, such as those using flask (CARLSON

and DUCKLOW, 1996) and mesocosm (LEBARON etal., 2001b) incubations, may identify direct effects,but might not accurately represent the in situ re-sponse of natural bacterioplankton communities tosystem-level enrichment. Conversely, although

large-scale comparative studies of multiple sys-tems (COLE, FINDLAY, and PACE, 1988; DEL GIOR-GIO and COLE, 1998) may identify differences inthe bacterioplankton community along an enrich-ment gradient, these data typically are integratedover large temporal and spatial scales. Conse-quently, they cannot be used to isolate the imme-diate and direct effect of system-level enrichmenton the bacterioplankton community alone.

The direct effect of system-level nutrient enrich-ment can be identified by combining the elementsof both large- and small-scale studies and con-ducting enrichment experiments on entire sys-tems. This approach has been implemented suc-cessfully in lakes (PACE and COLE, 2000) and theopen ocean (KOLBER et al., 1994), although thecharacteristically low water-residence time in tid-ally flushed systems (RASMUSSEN and JOSEFSON,2002) presents a challenge to this type of manip-ulation in most estuaries. An alternative is to useestuarine systems where enrichment gradients al-ready exist to simulate a large-scale nutrient en-richment experiment. For example, steep gradi-ents generated by point and nonpoint sources ofanthropogenic nutrient loading have been usedsuccessfully to evaluate the response of estuarinebacterioplankton to system-level nutrient enrich-ment (HOPPE, GIESENHAGEN, and GOCKE, 1998;REVILLA, IRIARTA, and ORIVE, 2000).

The Monie Bay component of the ChesapeakeBay National Estuarine Research Reserve in Mary-land (CBNERRMD) is an ideal system for this ap-proach to investigating the direct effect of system-level nutrient enrichment on estuarine systems.The research reserve is dominated by a shallow,semienclosed embayment and three creek systemsthat drain adjacent marshes and interact tidallywith bay waters. Differences in agricultural landuse among creek watersheds generate distinct pat-terns in nutrient enrichment both among and with-in the creek systems (CORNWELL, STRIBLING, andSTEVENSON, 1994; JONES, MURRAY, and STEVEN-SON, 1997). Additional predictable variability in nu-trient enrichment is introduced by the timing andnature of agricultural practices in each creek basin(CORNWELL, STRIBLING, and STEVENSON, 1994;JONES, MURRAY, and STEVENSON, 1997; FIELDING,2003). Thus, nutrient concentrations, land use, andsalinities in the three tidal creeks represent a broadrange of conditions on relatively small spatialscales, making Monie Bay an ideal system for com-parative investigations of the effect of agriculturalnutrient enrichment on salt-marsh communities


112 Apple, del Giorgio, and Newell

(JONES, MURRAY, and STEVENSON, 1997). In ad-dition, because this system exhibits a large rangeof conditions (i.e., nutrient concentrations, salinity,land use) over relatively small spatial scales, stud-ies conducted in this system are not hindered by theadditional variability typically imposed by differ-ences in basin or region-level processes and condi-tions (i.e., rainfall, climate, irradiance, temperature,atmospheric deposition of nutrients, etc.).

In this article we describe the use of Monie Bayas a natural experiment to investigate the effect ofsystem-level nutrient enrichment on estuarine bac-terioplankton communities. We begin by identifyingspatial and seasonal patterns of agricultural landuse, nutrient enrichment, and water column chem-istry within and among the three tidal creeks. Theeffect of system-level nutrient enrichment on thebacterioplankton community is subsequently ex-plored by (1) comparing agriculturally impacted vs.unimpacted tidal creeks; (2) comparing creeks dif-fering in terrestrial influence but experiencing sim-ilar enrichment; (3) evaluating changes along creekaxes from enriched headwaters to the relatively un-enriched open bay (OB); and (4) evaluating condi-tions before, during, and after pulsed inputs of ag-riculturally derived nutrients.

METHODS

Site Description

Monie Bay is a tidally influenced subestuary lo-cated on the eastern shore of Chesapeake Bay(388139300 N, 758509000 W). The reserve consists ofa relatively small (i.e., 1- to 2-kilometers wide and4-kilometers long) open bay (OB) and three tidallyinfluenced creeks varying in size and agriculturalland use (Figure 1). Monie Creek (MC) has thelargest of the three creek watersheds (45 squarekilometers), covering approximately 2.5 times and5 times more area than those of Little MonieCreek (LMC; 17.9 square kilometers) and LittleCreek (LC; 9.4 square kilometers), respectively.The linear reach of MC from headwaters to the OBis 6.5 kilometers, compared with 3.7 kilometers forLMC and 2.9 kilometers for LC. The creek chan-nels in these systems are the result of tidal scour-ing, with no significant fluvial input (WARD,KEARNEY, and STEVENSON, 1998). Monie Creekexperiences year-round inputs of fresh water,whereas LMC and LC have salinities driven en-tirely by tidal flushing from the OB and seasonalor episodic freshwater inputs occurring predomi-nantly in the spring or following major rain events

(JONES, MURRAY, and STEVENSON, 1997). MC andLMC are quite similar with respect to land usepatterns, with approximately 25% of each water-shed agriculturally developed and a similar pro-portion of watershed acreage attributed to marshand forest (Figure 1). As a result, MC and LMCare characterized by steep spatial gradients in nu-trient availability, with low-salinity regions expe-riencing elevated inputs of allochthonous nitrogenand phosphorus (JONES, MURRAY, and STEVEN-SON, 1997) that are ascribed to agricultural activ-ities (i.e., crop farming, livestock and poultry op-erations) within the watershed (CORNWELL, STRI-BLING, and STEVENSON, 1994). In comparison, LCwatershed is dominated by tidal marsh; approxi-mately one-third of the watershed is forested. Res-idential development in the LC watershed is min-imal and similar to that of other creeks (i.e., #3%),and there is almost no agricultural land use (i.e.,,1%).

The marsh macrophyte community of Monie Bayis dominated by Spartina spp. (S. alterniflora andS. patens), with Juncus roemerianus and Phrag-mites australis more prevalent in the upper marshthat experiences less frequent flooding (KEARNEY,STEVENSON, and WARD, 1994; STRIBLING andCORNWELL, 1997; WARD, KEARNEY, and STEVEN-SON, 1998). An exception is the upper reaches ofMC, which is characterized by a diverse freshwa-ter macrophyte community and greater abundanceof macrophytes that use C3 photosynthetic path-ways (JONES, MURRAY, and CORNWELL, 1997;STRIBLING and CORNWELL, 1997).

Experimental Design

Our investigation of the response of bacterio-plankton to system-level resource enrichmentcombines the manipulative aspect of a traditionalsmall-scale nutrient enrichment experiment (CA-RON et al., 2000; LEBARON et al., 2001b) with theecological relevance and larger spatial scale of insitu field observations. Using this approach, eachtidal creek is analogous to an individual treatmentin a small-scale manipulative experiment, where-by watershed characteristics (rather than the sci-entist) manipulate the environmental conditions ofinterest. For example, nutrient and DOM concen-trations and the source and quality of dissolvedand particulate organic seston are determined bythe extent of agricultural land use, dominant mac-rophyte cover, and watershed size. Tidal inunda-tion of each creek serves as an ‘‘inoculum’’ of OB



Figure 1. Monie Bay National Estuarine Research Reserve System with location and number of each sampling station (upperpanel) and proportion of each watershed attributed to one of four land-use categories (lower panel).

waters and associated bacterioplankton commu-nities. The resulting changes in the bacterioplank-ton community in each tidal creek relative to theOB are assumed to be a response to the environ-

mental conditions unique to each creek system.Analyses focused specifically on parameters relat-ed to resource regulation of bacterioplankton (i.e.,dissolved nutrients and DOM), although it is pos-



sible that top-down effects of grazers may alsohave an impact on estuarine bacterioplanktoncommunities (GONZALEZ, SHERR, and SHERR,1990; RIEMAN et al., 1990; DEL GIORGIO et al.,1996b). By sampling each creek system on the ebbtide, we were able to capture the metabolic re-sponse of bacterioplankton to these changing con-ditions, given that the tidal cycle is a comparabletime frame in which estuarine bacterioplanktonrespond to changing environmental conditions(PAINCHAUD et al., 1996). We used well-document-ed patterns in agricultural land use and nutrientavailability among and within the tidal creeks ofMonie Bay (CORNWELL, STRIBLING, and STEVEN-SON, 1994; JONES, MURRAY, and CORNWELL,1997) to define a range of comparisons (Figure 2);including horizontal (i.e., among creek), longitudi-nal (i.e., within creek), and temporal (i.e., seasonaland event based).

Horizontal Comparisons

Environmental conditions and biological param-eters in three tidal creek systems were comparedusing 2-year means (Table 1). Differences betweencreek systems were identified using analysis ofvariance (ANOVA), results from which were usedto classify the status of the three creeks (Figure 3)and form the basis for subsequent comparisons.For example, a comparison of LC and OB was usedto identify the effect of the marsh alone, specifi-cally the response of bacterioplankton to increasesin substrate enrichment in the absence of an in-crease in nutrient concentrations. This comparisonof LC and OB also provided a means by which com-parisons of LMC could be normalized for the effectof the marsh. A comparison of LMC and LC, ex-hibiting significantly different nutrient concentra-tions yet similar salinities, serves as nutrient en-richment and reference, respectively. Comparisonsbetween these two systems were used to identifythe effect of system-level nutrient enrichmentalone on estuarine bacterioplankton communities.Similarly, the two agriculturally impacted creeks(MC and LMC)—with similar ambient nutrientconcentrations but differences in DOM source andfreshwater inputs (JONES, MURRAY, and CORN-WELL, 1997)—were used to investigate the roles ofsubstrate source, quality, and quantity in medi-ating the effect of nutrient enrichment on bacter-ioplankton.

Longitudinal Comparisons

Longitudinal (i.e., within system) comparisonswere used to identify changes in environmentaland biological parameters along the creek axisfrom enriched conditions in the upper marsh to un-enriched conditions in the OB. Changes in envi-ronmental conditions and the bacterioplanktoncommunity along this axis were identified by re-gressions of salinity vs. environmental and micro-bial parameters of interest. This longitudinal ap-proach allowed the comparison of disparate con-ditions encountered among the creeks in this sys-tem (e.g., high vs. low nutrients), while alsorevealing the gradient between these extremesand tracking the corresponding shift in numerousenvironmental and biological parameters alongthe gradient.

Temporal Comparisons

Temporal variability of nutrient enrichmentadded a third dimension to the horizontal and lon-gitudinal comparisons described above (Figure 2),and can be in the form of pulsed enrichmentevents associated with the timing of agriculturalnutrient applications within the watershed (CORN-WELL, STRIBLING, and STEVENSON, 1994) or as-sociated with predictable seasonal changes in en-vironmental conditions (e.g., temperature, fresh-water inputs, irradiance). Pulsed nutrient inputswere used to evaluate the effect of system-level nu-trient enrichment on bacterioplankton by compar-ing pre- and postenrichment conditions, and pars-ing the 2-year dataset by season-identified generalseasonal effects. The interaction between system-specific (i.e., among creek) patterns and seasonwas identified using a full-factorial ANOVA withsystem (i.e., MC, LMC, LC, and OB), season(spring, summer, winter, fall), and their interac-tion as model effects.

Sample Collection and Estimates of BacterialAbundance and Production

We established 10 sites in the OB and tidal trib-utaries of Monie Bay (Figure 1). Two sites werelocated in both OB and LC, and three in each ofthe two agriculturally developed creeks. Stationswere located at intervals roughly proportional tothe total creek length and were selected to captureexisting gradients in nutrient concentrations andrelated water quality variables. In general, thesestations coincide with those used in previous mon-



Figure 2. Diagram of experimental approach used in Monie Bay. Comparisons were made in three dimensions, includinglongitudinal (transects along the creek axis), horizontal (comparisons among creek systems), and temporal (seasonal or event-based comparisons).

itoring and research projects (JONES, MURRAY,and CORNWELL, 1997; DEL GIORGIO, University ofMaryland Center for Environmental Science, per-sonal communication).

The 10 sites were visited monthly between April2000 and February 2002. Approximately 20 litersof subsurface (,0.5 meters) water were collectedin Nalgene HDPE carboys (Nalge Nunc Interna-tional, Rochester, NY) immediately following hightide and transported back to the laboratory for fil-tration. Water temperature, salinity, Secchi depth,and water column depth were recorded at eachsite. Upon return to the laboratory, a small sub-sample was removed from each carboy for deter-mining total bacterioplankton production andabundance, and concentrations of inorganic nutri-ents, and dissolved organic carbon (DOC). Bacte-rial production (BP) was estimated from the up-take of 3H-leucine according to the centrifugationmethod of SMITH and AZAM (1992). Rates of leu-cine uptake were measured in all unfiltered water

samples to gather an estimate of the total com-munity production, which includes free-living andattached bacterioplankton. Estimates of filteredbacterial production were determined by gentlypassing several liters of sample water through anAP15 Millipore (Billerica, MA) filter (;1 micro-meter) using a peristaltic pump, then incubatingthe sample in the dark at in situ field temperature.There were three measurements of leucine uptakein the filtered fraction during the incubation (at 0,3, and 6 h), and these individual measurementswere averaged to obtain a mean rate of bacterialleucine uptake for the incubation period. Rates ofleucine uptake were converted to rates of carbonproduction assuming a conversion factor of 3.1 ki-lograms C mol leu21 (KIRCHMAN, 1993). Bacterio-plankton abundance (BA) was determined on livesamples using standard flow-cytometric tech-niques and the nucleic acid stain SYTO-13 (DEL

GIORGIO et al., 1996a).



Table 1. Nutrient concentrations, biological parameters, and watershed characteristics for the three tidal creeks and open bay.Table values are derived from 2-year means 6 SE (n).

MC LMC LC OB n

Depth (m)TDN (mM)TDP (mM)DON (mM)NH4

1 (mM)

3.3 6 0.2 (29)40.6 6 2.1 (65)0.7 6 0.08 (65)

36.7 6 1.7 (65)2.3 6 0.314 (65)

3.1 6 0.2 (25)40.1 6 2.8 (58)0.8 6 0.12 (58)

35.6 6 1.9 (58)3.0 6 0.428 (58)

2.0 6 0.1 (18)26.8 6 1.6 (38)0.2 6 0.03 (38)

21.6 6 1.9 (38)2.1 6 0.2 (38)

2.2 6 0.1 (14)28.1 6 1.7 (40)0.3 6 0.03 (40)

19.8 6 1.8 (40)1.7 6 0.178 (40)

86201201201201

NOx (mM)PO4

32 (mM)SalinityDOC (mg L21)DOC : TDN

4.5 6 0.859 (65)0.2 6 0.05 (65)6.9 6 0.4 (75)

11.5 6 0.5 (71)24 6 2 (62)

4.8 6 1.045 (58)0.3 6 0.065 (58)9.9 6 0.3 (62)8.9 6 0.4 (59)20 6 1 (55)

3.1 6 0.64 (38)0.2 6 0.1 (37)

11.6 6 0.3 (38)7.7 6 0.3 (36)27 6 2 (36)

6.6 6 1.299 (40)0.0 6 0.006 (40)

12.1 6 0.3 (42)6.0 6 0.2 (39)20 6 1 (37)

201200217205190

TDN : TDPa350*1

BA (106 cells mL21)BP (mg C L21 h21)

103 6 22 (60)20 6 0.7 (43)

11.8 6 0.7 (74)1.8 6 0.1 (59)

84 6 8 (55)17 6 0.7 (35)

13.3 6 1.0 (62)2.6 6 0.2 (48)

170.0 6 27 (34)15 6 1 (19)

12.8 6 1.4 (35)1.5 6 0.1 (29)

130 6 15 (35)12 6 0.9 (22)

11.5 6 1.0 (42)1.1 6 0.1 (33)

184119213169

Filtered BP (mg C L21 h21)3 1.0 6 0.1 (59) 1.4 6 0.1 (48) 1.0 6 0.2 (29) 0.6 6 0.1 (33) 169% filtered BP2

Watershed size (km2)Agriculture4

55.6 (59)4523

53.8 (48)17.925

66.7 (29)9.4

,1

54.5 (33)72.316

169

1 Specific absorbance at 350 nm 3 103.2 BP for the AP15 filtered fraction.3 Percentage of total BP attributed to the AP15 filtered fraction.4 Percentage of agricultural land use within each watershed. The open bay watershed is composed of adjacent marshes andthe watershed from each creek.MC 5 Monie Creek, LMC 5 Little Monie Creek, LC 5 Little Creek, OB 5 open bay, TDN 5 total dissolved nitrogen, TDP5 total dissolved phosphorus, DON 5 dissolved organic nitrogen, NOx 5 NO3

2 1 NO22, DOC 5 dissolved organic carbon, BA

5 total bacterial abundance, BP 5 total bacterioplankton production.

Nutrients and Other Analyses

Filtered samples for DOC analysis were acidi-fied with 100 microliters of 1 normal phosphoricacid and held at 48C until analysis. DOC contentwas determined with a Shimadzu high-tempera-ture catalyst carbon analyzer (Shimdazu Corpo-ration, Kyoto, Japan) (SHARP et al., 1995). Samplesfor nutrient analyses were filtered through aWhatman GF/F filter (Whatman Inc., Clifton, NJ)and frozen at 2258C for later analysis of phos-phate (i.e., PO4

32, soluble reactive phosphorus), ni-trite (NO2

2), and nitrate (NO32) following STRICK-

LAND and PARSONS (1972); total dissolved nitro-gen (TDN) and total dissolved phosphorus (TDP)following VALDERRAMA (1981); and ammonium(NH4) following WHITLEDGE et al. (1981). Dis-solved organic nitrogen (DON) was determined asthe difference between TDN and DON compo-nents. Absorbance of DOC was determined on GF/F filtered samples by performing absorbance scans(290–700 nanometers) using a Hitachi U-3110spectrophotometer (Hitachi Corporation, Tokyo,Japan) and either 1- or 5-centimeter quartz cu-vettes, depending upon the concentration of col-ored dissolved organic matter (CDOM). Absorptiv-

ity at 350 nm (a350) was used as an index of CDOMconcentrations. Specific absorbance (a350*) was de-termined by dividing a350 by ambient DOC concen-trations (MORAN, SHELDON, and ZEPP, 2000; HU,MULLER-KARGER, and ZEPP, 2002). Chlorophyll awas determined with standard methods using aTurner 10-AU fluorometer (Turner Designs, Sun-nyvale, CA) (STRICKLAND and PARSONS, 1972).

Land Use and Watershed Designations

Land use within the watersheds of MC, LMC,and LC was classified as residential, agricultural,marsh, or forest (ANDERSON et al., 1976) usinggeo-referenced satellite imagery provided by theMaryland Department of Natural Resources. Thewatershed for each of the three tidal creeks wasidentified based on boundaries established by theUS Geological Survey for first- and second-orderstreams within the larger Monie drainage basin.The relative proportion of land use for each creeksystem was calculated using the watershed desig-nations identified above and land use polygon sizerelative to the entire watershed area (LEE et al.,2000).



Figure 3. Horizontal comparisons of 2-year means among systems. For each parameter, bar height represents the magnitudeof the 2-year mean. Means that are statistically similar share the same bar height. (ANOVA and Tukey-Kramer HSD, a50.05).Parameters are defined in Table 1.

Statistical Analyses

All statistical analyses, including standard leastsquares regressions, one- and two-way ANOVA,and principal components analysis (PCA) were per-formed using the JMP 5.0.1 statistical softwarepackage (SAS Institute, Inc., Cary, North Carolina).The entire composite dataset was used for all sta-tistical analyses, except ANOVAs comparing sys-tem-specific means (Figure 3, Table 2). In these in-stances, each system was characterized by valuesobserved at the uppermost two sites in each creek(i.e., sites 3 and 4, 5 and 6, and 8 and 9 in LC, LMC,

and MC, respectively). Because of significant differ-ences in rainfall and salinity among years, regres-sion statistics for longitudinal comparisons with sa-linity as the independent variable (i.e., Table 3)were performed on data from year 1 only.

RESULTS

The three tidal creeks of Monie Bay differ withrespect to agricultural land use, nutrient and DOCconcentrations, salinity, and rates of bacterial pro-duction (Table 1). We observed significant differ-ences in these measured parameters among creek



Table 2. Probability values from two-way ANOVA tests with system and season as model effects.

Parameter

Model Effects

System Season Interaction n

TemperatureTotal dissolved nitrogenTotal dissolved phosphorusDissolved organic nitrogen

ns,0.0001,0.0001,0.0001

,0.0001,0.0001

nsns

nsnsnsns

176162162162

AmmoniumNitrate and nitritePhosphateDissolved organic carbon

0.0010.05ns

,0.0001

,0.0001,0.0001

ns,0.0001

0.0030.05ns

,0.0001

162162160166

SalinitySpecific absorbanceTotal bacterial productionTotal bacterial abundance

,0.0001,0.0001,0.0001

ns

,0.00010.0010.01

,0.0001

0.02nsnsns

17697

137172

ns 5 not significant (p . 0.05).

systems and along longitudinal creek transects(Figure 3). In addition, sampling over the 2-yearduration of this study revealed considerable sea-sonal fluctuations in these parameters amongcreek systems. Our evaluation of this overlappingvariability in Monie Bay and the correspondinghorizontal, longitudinal, and seasonal comparisons(e.g., Figure 2) are described below.

Horizontal Comparisons among Creeks

In general, dissolved nutrient concentrationswere higher in the two creeks with agriculturallydeveloped watersheds and similar in LC and OB(Table 1). A statistical comparison of 2-year means(Figure 3; ANOVA; Tukey-Kramer honest signifi-cant difference [HSD]; a50.05) indicates thatTDN, TDP, and DON were significantly higher inMC and LMC relative to both LC and OB. Therewere no statistical differences in dissolved ammo-nium and phosphate among the creek systemswhen 2-year means were considered, although con-centrations in MC were significantly higher thanthose of OB. We observed considerable seasonalvariability in nitrate, resulting in no significantdifferences among systems in this parameter (datanot shown). In addition, dissolved nutrient stoi-chiometry (N : P; Table 1) in MC and LMC sug-gests that these systems are disproportionately en-riched with phosphorus relative to nitrogen whencompared with LC and OB. Mean salinity de-creased with increasing watershed size, and therewas a consistent hierarchy among the creeks(MC,LMC,LC,OB; Table 1, Figure 3). Both col-ored and total DOC were inversely related to sa-linity, with highest values of DOC and CDOM in

MC and lowest in OB and similar among-systemhierarchy to that of salinity (i.e., MC.LMC.LC.OB). The pattern in CDOM mirrored that ofsalinity (Figure 3), with significantly higher a350*values in MC relative to all other systems and sim-ilar values when LMC and LC, as well as LC andOB, were compared.

Longitudinal Patterns within Creeks

We explored changes in water column chemistryand biology along the estuarine axis using regres-sions of salinity vs. other measured parameters(Table 3). All parameters decreased to varying de-grees along the creek axis. There were significantdecreases in DOC, phosphate, TDN, TDP, andDON in both MC and LMC, and a significant de-crease in total BP along the creek axis in LMC.The concurrent decrease of inorganic nutrient con-centrations and BP in LMC is illustrated in Figure4. There were no significant changes in nutrientsalong the axis of Little Creek, although the trendof decreasing BP was significant (Table 2).

Seasonal and Temporal Patterns in Monie Bay

Dissolved nutrient concentrations were highlyvariable throughout the 2-year sampling period,with seasonal maxima of both TDP and TDN inApril and July 2000 and April and September 2001(Figure 5). In April 2000, TDN peaked at 90 mMin both MC and LMC, and TDP was 2 and 4 mMin MC and LMC, respectively. A second peak oc-curred in July 2000, with TDN concentrations of73 and 50 mM in MC and LMC and TDP concen-trations of 3 and 1 mM, respectively. In 2001, April



Tab

le3.

Reg

ress

ion

stat

isti

csfo

rth

ere

lati

onsh

ipbe

twee

nsa

lini

tyan

dph

ysic

alan

dbi

olog

ical

para

met

ers

for

the

uppe

rmos

ttw

osi

tes

inea

chcr

eek

(yea

ron

eon

ly;n

512

).

Par

amet

er

Mon

ieC

reek

Slo

per2

Fra

tio

p

Lit

tle

Mon

ieC

reek

Slo

per2

Fra

tio

p

Lit

tle

Cre

ek

Slo

per2

Fra

tio

p

Tot

aldi

ssol

ved

nitr

ogen

Tot

aldi

ssol

ved

phos

phor

usD

isso

lved

orga

nic

nitr

ogen

Am

mon

iaN

itra

tean

dni

trit

e

24.

82

0.2

23.

12

0.3

21.

5

0.37

0.45

0.39

0.12

0.28

7.1

9.9

7.7

1.7

4.6

0.02

0.00

90.

02ns 0.

05

25.

12

0.1

23.

62

0.4

21.

1

0.52

0.44

0.58

0.16

0.22

10.9 7.7

13.9 1.8

2.9

0.00

80.

020.

004

0.2

0.12

21.

50.

032

0.4

20.

32

0.8

nr nr nr 0.14 nr

0.33

0.83

0.05

1.7

0.45

ns ns ns 0.2

nsP

hosp

hate

Dis

solv

edor

gani

cca

rbon

Spe

cific

abso

rban

ce(a

350*

)T

otal

bact

eria

lpr

oduc

tion

20.

12

1.5

20.

001

20.

2

0.50

0.56

0.48

0.16

12.0

15.1

18.2 3.6

0.00

50.

002

0.00

040.

075

20.

12

1.2

20.

001

20.

8

0.52

0.79

0.33

0.89

10.9

38.4

11.0

71.6

0.00

80.

0001

0.00

30.

0001

20.

012

1.0

20.

0004

20.

4

0.26

0.48

0.15

0.51

3.1

9.4

2.9

10.3

0.1

0.01

0.1

0.00

9F

ilte

red

bact

eria

lpr

oduc

tion

Tot

alba

cter

ial

abun

danc

e2

0.1

—nr nr

1.6

—0.

23ns

20.

6—

0.80 nr

32.1

—0.

0005

ns2

0.3

—0.

34 nr5.

1—

0.05

ns

nr5

nore

lati

onsh

ip(r

2,

0.1)

,ns

5no

tsi

gnifi

cant

(p.

0.05

).

concentrations were 81 and 118 mM for TDN and2 and 4 mM for TDP in MC and LMC, respectively.The summer peak in nutrients occurred later in2001, with TDN concentrations of 42 and 62 mMand TDP concentrations of 1 and 6 mM in Septem-ber in MC and LMC, respectively. TDN and TDPconcentrations in LC and OB did not exhibit thesame seasonal pattern of enrichment. Despite theconsiderable interannual variability, the patternof nutrient concentrations among creeks persistedsuch that TDP and TDN were always higher in theagriculturally influenced creeks relative to LC andOB for all dates sampled. Bacterial production wasalways highest in LMC, higher in all creeks thanin the OB, and generally followed seasonal pat-terns in temperature (Figure 5, lower panel).

We evaluated differences in 2-year meansamong seasons using ANOVA and Tukey-KramerHSD (Figure 6). Spring was characterized by lowersalinity and higher TDN and NOx (NO2

2 1 NO32)

concentrations. There was no significant differencein TDP or DON among seasons, and phosphateconcentrations were similarly elevated in all sea-sons but winter. Salinity increased throughout theseasons, with lowest measurements in spring andhighest measurements in winter, and was gener-ally mirrored by DOC and CDOM concentrations.Temperature and BP followed a similar seasonalpattern, with highest values in summer and lowestvalues in winter, and BA was significantly higherin the spring.

Given the overlapping spatial and seasonal var-iability in Monie Bay, we investigated the inter-action of these factors by conducting a two-wayANOVA with system (creek) and season as modeleffects (Table 2). We observed significant interac-tions between season and system when salinity,DOC, salinity, NH4

1, and NOx were considered.These interactions correspond to disproportionate-ly elevated DOC concentrations and reduced salin-ity in MC in the spring, lower NOx concentrationsin LC in the spring relative to other systems, anddisproportionately elevated NH4

1 concentrationsin LMC in the spring.

Principal Components Analysis

Principal components analysis identified twocomposite variables (hereafter PC1 and PC2) thatexplained 75.4% of the variability within the com-posite dataset (n 5 160), with 47.6% and 27.8%attributed to PC1 and PC2, respectively. PC1 hadhigh factor loadings (eigenvectors .0.8) for DOC,



Figure 4. Transect of ambient nutrient concentrations (total dissolved phosphorus [TDP] and total dissolved nitrogen [TDN])and total bacterial production (BP) along the axis of agriculturally impacted Little Monie Creek (LMC) and open bay (OB;each point represents 2-year mean 6 SE, n521).

TDN, TDP, and DON and was negatively correlat-ed with salinity, whereas PC2 was strongly cor-related with temperature. The distribution of sam-pling events from MC and LMC on PC1 and PC2(Figure 7, upper panel) identifies the similaritiesbetween these systems and indicates that vari-ability in these systems is dominated by freshwa-ter delivery of dissolved nutrients and organicmatter, and that they experience similar nutrientloading dynamics. In contrast, the negative corre-lation of sampling events from LC and OB withPC1 (Figure 7, lower panel) indicates higher salin-ities (i.e., reduced freshwater inputs) and minimalnutrient loading. Temperature and/or seasonal ef-fects explain most of the variability in these sys-tems, as evidenced by the distribution along PC2.

We explored seasonal patterns in water columnchemistry using PCA and the same composite da-taset, parsed by season in Figure 8. Samples col-lected in spring were positively correlated withPC1 and negatively correlated with PC2, indicat-ing elevated concentrations of dissolved nutrientsand DOC, and lower temperatures during this sea-son. Samples collected during summer and fallwere positively correlated with PC2 (associated

with higher water temperatures) and had a simi-lar distribution along PC1. Samples collected inwinter were negatively correlated with both PC1and PC2.

Bacterial Production and Abundance

Two-year means of total bacterial productionwere highest in the agriculturally developedcreeks and lowest in LC and the OB(LMC.MC.LC.OB; Table 1). Bacterial produc-tion in LMC (2.6 6 0.2 micrograms C liter21 h21)was significantly higher than that of LC and OB(1.5 6 0.1 and 1.1 6 0.1 micrograms C liter21 h21,respectively). Although BP in MC (1.8 6 0.2 mi-crograms C liter21 h21) was higher than that of LC,this difference was not significant (Figure 3). BPin both MC and LMC was significantly higherthan that of OB. Bacterial production in the fil-tered fraction was always lower than that of totalbacterial production. The contribution of the fil-tered fraction to total production ranged from ap-proximately 54% in MC, LMC, and OB to 67% inLC (Table 1). Bacterial abundance was higher inLMC and LC and lower in MC and OB (Table 1),



Figure 5. Two-year seasonal variability in total dissolved phosphorus (TDP), total dissolved nitrogen (TDN), total bacterialproduction (BP), and temperature (TEMP) among the sub-systems of Monie Bay.

although these differences were not significant.Bacterial production decreased from the upper es-tuary to the OB in all creeks, with the largest andmost significant change in LMC (Figure 4, Table3). A similar but weaker trend was observed in thefiltered fraction in both LMC and LC. There wereno significant changes in BA along creek axes.

DISCUSSION

Monie Bay exhibits a diverse range of environ-mental conditions and watershed characteristicswithin one small estuarine system. Spatial andtemporal patterns in nutrient concentrations and

organic matter loading among the three tidalcreeks and the OB create conditions ideal for theuse of this system as a natural experiment to in-vestigate the effects of system-level nutrient en-richment. Our study in Monie Bay revealed con-sistent relationships between agricultural landuse, ambient nutrient concentrations, freshwaterinput, and rates of bacterial production. The bac-terioplankton community responds positively tosystem-level nutrient enrichment, although thisresponse appears to be mediated by nutrient andorganic carbon delivery associated with patternsin freshwater input to these tidal creeks.



Figure 6. Comparisons of seasonal means for environmental and biological parameters measured over the 2-year samplingperiod. For each parameter, bar height represents the magnitude of the 2-year mean. Means that are statistically similarshare the same bar height. (ANOVA and Tukey-Kramer HSD, a50.05). Parameters are defined in Table 1.

Patterns in Nutrient Enrichment

Nutrient enrichment in the creeks of Monie Bayis a function of both short- and long-term nutrienttransport mechanisms, including baseline inputsof nitrogen from groundwater and pulsed inputs ofnitrogen and phosphorus associated with fertilizerapplication and rainfall events. These loadingsgenerate distinct patterns of nutrient enrichmentthat are apparent throughout the year both withinand among creek systems. We observed a persis-tent hierarchy of ambient nutrient concentrations

among the creeks (MC.LMC.LC) during allmonths sampled as well as when 2-year meanswere considered. Despite significant differences insalinity (Figure 3) and freshwater inputs (JONES,MURRAY, and CORNWELL, 1997), MC and LMCare remarkably similar with respect to the dynam-ics of nutrient delivery (Figures 5 and 7). Mea-sured nutrient concentrations among the threecreeks exhibited an identical pattern and similarmagnitude to those observed by JONES, MURRAY,and CORNWELL (1997), reaffirming the robust



Figure 7. Principal components analysis of each sampling event with loadings on PC1 and PC2. Sampling events are sepa-rated by system: Monie Creek and Little Monie Creek (upper panel) and Little Creek and the open bay (lower panel; n5160).

nature of spatial patterns of nutrient enrichmentin Monie Bay.

Short time-scale inputs, such as those associatedwith fertilizer application within the watersheds,have an episodic effect on the enrichment of thesystem—a phenomenon that has been well docu-mented in other agriculturally developed water-sheds of this region (LOWRANCE et al., 1997; STAV-ER and BRINSFIELD, 2001). The timing of these pe-riods of enrichment (Figure 5) coincides with fer-tilizer application schedules in the Monie Baywatershed, where chicken manure and/or liquidurea are applied in late March to early April, fol-lowed by the application of liquid urea in June

(WILLIAMS, Sommerset County Agricultural Ex-tension, personal communication), suggesting thatthe periodic acute enrichment of this system isdriven by fertilizer application to fields within thedrainage basins and the subsequent transport ofwater and associated nutrients into the tidalcreeks during rain events (SPEIRAN, HAMILTON,and WOODSIDE, 1998; NORTON and FISHER,2000). CORNWELL, STRIBLING, and STEVENSON

(1994) and JONES, MURRAY, and CORNWELL

(1997) observed a similar timing of maxima in nu-trient concentrations and also attributed these toagricultural nutrient loading associated with fer-tilizer and manure applications. The negative



Figure 8. Principal components analysis of each sampling event with loadings on PC1 and PC2. Sampling events are iden-tified by season (n5160).

loading of salinity and positive loading of dissolvednutrients on PC1 indicate that freshwater inputsdrive most of the nutrient delivery to these sys-tems. Concurrent enrichment of the two agricul-turally developed creeks (MC and LMC) with bothTDN and TDP (Figure 5) implicates overland flowfrom rainfall events as a loading mechanism, as itis well documented that phosphorus is transportedwith sediment via storm flow and/or erosionalevents (NORTON and FISHER, 2000). In addition,subsurface transport may also be responsible forless episodic inputs of phosphorus to these sys-tems. SIMS, SIMARD, and JOERN (1998) observedenvironmentally significant inputs of phosphorusvia subsurface flow in systems where excessive useof organic wastes increased soil phosphorus con-centrations well above crop requirements. Long-term application of phosphorus-rich chicken ma-nure to fields within MC and LMC may have re-sulted in concentrations approaching the sedimentadsorption maxima (SIMS and WOLF, 1994), a con-sequence of which is an increase in the equilibriumconcentration in subsurface waters and leaching ofphosphorus into adjacent aquatic systems (SIMS

and WOLF, 1994; SIMS, SIMARD, and JOERN,1998).

The importance of freshwater inputs in drivingnutrient delivery in this system suggests thatthere will also be distinct patterns in nutrient en-richment among seasons. We observed significant-

ly lower salinity and significantly higher concen-trations of TDN and NOx in the spring. The lackof a similar pattern in TDP and PO4

32 suggeststhat the delivery of nitrogen at this time was driv-en by a general increase in freshwater input andnot necessarily by overland flow related to episodicstorm events. Freshwater delivery to the creeks ofMonie Bay is lowest in winter, as evidenced by el-evated salinity and reduced PO4

32, NOx, TDN,DOC, and CDOM at this time. For the most part,these seasonal effects are independent of the spa-tial patterns observed among the creeks, althoughthere were certain conditions under which therewas significant interaction of these effects (Table2). We observed the strongest interaction in thespring, with disproportionately low salinities andelevated DOC in MC and disproportionately ele-vated ammonium in LMC. The pattern in MC canprobably be attributed to a larger drainage basinmore effectively delivering spring rainfall andstored organic matter from macrophyte senescencethe previous fall. All systems except LC were dis-proportionately loaded with NOx in the spring,suggesting that temporally based (i.e., not systembased) comparisons are more appropriate for iden-tifying the effects of nitrate in Monie Bay, and thatLC is indeed more pristine with respect to the im-pact of agricultural nutrients.

The transition from elevated nutrient concentra-tions in the upper estuary, where agricultural



development is greatest (Figure 1), to lower con-centrations near the bay (Table 3, Figure 4) wascorroborated by other investigators (CORNWELL,STRIBLING, and STEVENSON, 1994; JONES, MUR-RAY, and CORNWELL, 1997). JONES, MURRAY, andCORNWELL, (1997) specifically report a doubling ofnitrogen and phosphorus concentrations along atransect from the OB to headwaters of LMC. Thesepatterns observed in multiple studies clearly sug-gest that nutrients from agricultural land use en-ter each creek upstream, and are measurably di-luted or consumed as they pass downstream intothe marsh and are subjected to tidal mixing.

In addition to patterns of acute nutrient enrich-ment associated with agricultural practices, we ob-served significant and persistent differences in nu-trient concentrations among the creeks duringmonths of little or no fertilizer application (Figure5). This indicates that acute periodic inputs aug-ment a more chronic, background level of inputsfrom contaminated groundwater and surficialaquifers that have been infiltrated by agricultur-ally derived nutrients (WEIL, WEISMILLER, andTURNER, 1990; SPEIRAN, HAMILTON, and WOOD-SIDE, 1998). This nutrient delivery provides a rel-atively constant, low-level input via base flow thatreflects long-term (i.e., 5–20 y) agricultural landuse in the watershed (WEIL, WEISMILLER, andTURNER, 1990; SPEIRAN, HAMILTON, and WOOD-SIDE, 1998). As a result, despite extensive vari-ability in nutrient concentrations throughout theyear, monthly and annual nutrient concentrationsin the impacted creeks are always significantlyhigher than those of the reference creek (Figures3 and 5). Additional evidence of the long-term ef-fects of agricultural practices on the tidal creeks ofthis system is revealed by dissolved nutrient stoi-chiometry. JONES, MURRAY, and CORNWELL

(1997) report extremely low N : P ratios in LMC,and attribute this to the high phosphorus contentof chicken manure (SIMS and WOLF, 1994) pro-duced and applied in the LMC watershed. Despitethe small areal coverage of poultry farms locatedin the LMC drainage basin (i.e., 0.9% of the entireMonie Bay watershed), these facilities account forthe majority of nitrogen and phosphorus inputs tothe Monie Bay system (81% and 68%, respectively;JONES, MURRAY, and CORNWELL, 1997). It is clearthat this is a persistent if not long-term effect, giv-en that we observed the same pattern in N : P ra-tios among the tidal creeks (lowest in LMC; Table1) almost a decade after the original 1994 fieldwork of JONES, MURRAY, and CORNWELL (1997).

Response to System-Level Enrichment

Nutrient enrichment of the tidal creeks in MonieBay has a pronounced effect on the productivityand functioning of these systems, driving patternsof marsh macrophyte productivity and biomass(JONES, MURRAY, and CORNWELL, 1997), as wellas sediment biogeochemistry and nutrient cycling(CORNWELL, STRIBLING, and STEVENSON, 1994;STRIBLING and CORNWELL, 2001). Our study re-vealed that bacterioplankton also respond to sys-tem-level nutrient enrichment, although this re-sponse differs among the creek systems and ap-pears to be modulated by the interaction of variousenvironmental factors. Elevated BP and DOC inLC relative to OB suggest that bacterioplanktonrespond positively to marsh-derived increases inorganic matter supply. We observed a similar pat-tern in LMC, where inputs of agriculturally de-rived nutrients combine with marsh-derived or-ganic matter to produce the highest rates of BPrecorded among the tidal creeks of Monie Bay. De-spite comparable nutrient and organic matter en-richment in MC relative to LMC, we did not ob-serve elevated rates of BP in this system, suggest-ing that additional factors mediate the response ofbacterioplankton to system-level enrichment. Wepredict that the muted response to enrichment ob-served in MC is driven by allochthonous inputs oflower-quality, terrestrially derived DOM.

Effect of the Marsh

Higher bacterial abundance and production inLC relative to the OB (Table 1) suggest that themarsh environment itself has a positive effect onthe bacterioplankton community. Similar trends ofincreased production and abundance have been ob-served in other temperate estuaries (HOCH andKIRCHMAN, 1993; GOOSEN et al., 1997; REVILLA,IRIARTA, and ORIVE, 2000) and tidal creeks of theChesapeake Bay (SHIAH and DUCKLOW, 1995) andhave generally been attributed to inputs of labilemarsh detritus (BANO et al., 1997; REITNER, HER-ZIG, and HERNDI, 1999). Our observation of higherrates of BP and DOC concentrations in LC relativeto OB (Table 1), consistently higher BP in LC vs.OB at all sampling events (Figure 5), and a sig-nificant increase in both DOC and BP along theaxis of LC (Table 3), corroborates these studiesand further suggests that there is a positive effectof marsh detritus on BP. Similar nutrient concen-trations between LC and OB (Table 1, Figure 3)indicate that these increases in BP are driven by



changes in the quality and quantity of DOM andPOM substrates associated with natural marshprocesses (SHIAH and DUCKLOW, 1995; GOOSEN etal., 1997), rather than an effect of nutrients alone.Thus, elevated rates of BP observed in agricultur-ally impacted marsh systems are most likely driv-en by a combination of the direct effect of anthro-pogenic nutrient enrichment (JONES, MURRAY,and CORNWELL, 2000) and the positive effect ofnatural marsh processes, although the effect of themarsh is probably minimal relative to that of sys-tem-level nutrient enrichment (SCUDLARK andCHURCH, 1989).

Effect of Enrichment: Little Monie Creek vs. LittleCreek

Our comparison of LMC and LC was used to iso-late the effect of system-level nutrient enrichmenton bacterioplankton, an approach that relies onthese systems being comparable in all aspects oth-er than agricultural nutrient loading. As part oftheir 2-year study of these creeks, JONES, MUR-RAY, and CORNWELL (1997) concluded that theoverall similarity in physical parameters—coupledwith differences in watershed practices—makesthese creeks directly comparable and provides anexcellent study area to assess the impact and ul-timate fate of agricultural nutrients in brackishmarsh systems. Given that LMC and LC are ad-jacent watersheds (Figure 1), it is unlikely thatlarge spatial-scale processes (i.e., climate, precipi-tation, atmospheric deposition of nutrients) willcontribute to differences between these systems,and we predict that differences in nutrient concen-trations between these systems are predominantlya function of agricultural land use, extent of marshacreage, and/or watershed size and hydrology(NORTON and FISHER, 2000).

Watershed size does not appear to have an ef-fect. Although the watershed size of LMC is onlytwice that of LC (Table 1), LMC has higher TDPand PO4

32 during all months sampled by a factorof 4.5 and 6.5, respectively (data not shown).JONES, MURRAY, and CORNWELL (1997) reportsimilar findings, with phosphorus concentrationsin LMC being fourfold higher than those of LC.Thus, based on watershed size, LMC is dispropor-tionately enriched with phosphorus relative to LC.

With respect to dissolved nitrogen, JONES, MUR-RAY, and CORNWELL (1997) report and we ob-served concentrations twofold to threefold higherin LMC than LC. This difference suggests that

LMC is not as enriched with nitrogen as withphosphorus, although it is more likely that TDNconcentrations in LC are influenced by the inputsof nitrogen-enriched groundwater (WEIL, WEIS-MILLER, and TURNER, 1990; SPEIRAN, HAMILTON,and WOODSIDE, 1998) or the influx of nitrogen-laden waters from the OB during periods of nitro-gen loading to the entire system. For example,when nutrient-rich water from MC and LMC istransported to the OB during ebb tide, it is thenintroduced to LC via tidal interactions. This phe-nomenon can be observed in the concurrent peaksof TDN in all three tidal creeks (Figure 5). In theirevaluation of nutrient concentrations over thecourse of a tidal cycle, JONES, MURRAY, andCORNWELL (1997) found the highest nutrient con-centrations in LC at high tide, further suggestingdelivery of dissolved nitrogen from the OB. Suchenrichment of LC may lead to a smaller apparentdifference between annual TDN concentrations ob-served in LMC and LC, inaccurately suggestingthat LMC may not be disproportionately enrichedwith nutrients. Phosphorus does not experiencethe same effect as nitrogen, as it is transported inthe particulate phase during storm and runoffevents (NORTON and FISHER, 2000).

Given the proportion and extent of marsh acre-age in the LC watershed relative to that of LMC(63% vs. 30%, respectively; Figure 1), it is also pos-sible that natural marsh processes may contributeto differences in nutrients between these systems.CORNWELL, STRIBLING, and STEVENSON (1994)and STRIBLING and CORNWELL (2001) report asignificant effect of the marsh on nutrient budgetsin Monie Bay. The authors observe a decrease innutrient concentrations over the course of thegrowing season, attributing the decrease to con-sumption of nitrogen and phosphorus by marshmacrophytes and loss of nitrogen via sediment de-nitrification. These studies also report a signifi-cant contribution of the marshes to water-columnNH4

1 via sediment ammonification. If the exten-sive marsh acreage in LC is a significant sink forwater-column nitrogen and phosphorus—and thuscontributes to differences in nutrient concentra-tions between these two systems—then it shouldalso be a source of ammonium. However, compar-isons of 2-year means (Table 1), creek transects(Table 3), and data from JONES, MURRAY, andCORNWELL (1997) reveal no such enrichment ofLC with ammonium, and we conclude that elevat-ed nutrient concentrations in LMC are drivenexclusively by agricultural inputs, with only



negligible effects attributed to catchment size andextent of marsh coverage.

Biological Response to Enrichment

JONES, MURRAY, and CORNWELL (1997) identi-fied an effect of agricultural nutrient enrichmentamong the tidal creeks of Monie Bay, with elevat-ed plant biomass, tissue nutrient concentrations,and water column chlorophyll a in LMC relativeto that of LC. These changes in the macrophytecommunity were correlated with rainfall and as-sociated runoff events, further indicating that nu-trient delivery to this system is derived from ag-ricultural practices. The positive effect of enrich-ment on marsh macrophytes and phytoplanktonwas reflected in the positive relationship betweenBP and system-level enrichment in LMC, indicat-ing that elevated productivity in LMC is a robustand consistent pattern that can be observed onmany levels of biological organization. Despite con-siderable interannual variability in BP and nutri-ent concentrations, we observed consistently high-er rates of bacterial production in LMC relative toLC throughout the year (Figure 5), when 2-yearmeans from these systems were compared (Figure3), and when the nutrient-enriched upper estuaryof LMC was compared with sites nearer the OB(Figure 4).

In addition, although we did not observe a sig-nificant difference in bacterioplankton abundancebetween these two creeks (Table 1), cell-specificproduction (i.e., bacterial production per individualcell) in LMC was significantly higher than that ofall other systems (Tukey-Kramer HSD; a50.05;p,0.0001; n5137; Table 1, Figure 3). We predictthat the observed increases in cell-specific produc-tion in LMC were associated with nutrient-drivenincreases in the growth and metabolism of individ-ual cells within the assemblage, such that small,dormant, or slow-growing cells became more activeand larger in direct response to enriched condi-tions (DEL GIORGIO and SCARBOROUGH, 1995;CHOI, SHERR, and SHERR, 1999). A comparison ofLC and LMC revealed that total BP in LC wasdominated by the filtered fraction (,1 micrometer)relative to LMC (67% vs. 54%, respectively). Thisdifference in filtered vs. total BP indicates a de-crease in the relative abundance of small, free-liv-ing cells in LMC and suggests a shift of bacterio-plankton to a particle-attached state associatedwith elevated POM in this system (JONES, MUR-RAY, and CORNWELL, 1997) or an increase in the

abundance of larger, more rapidly growing cellsthat are then retained in the AP15 filter (GASOL

and DEL GIORGIO, 2000). Thus, the change in totalbacterial production observed in LMC not onlyrepresents a general increase in bacterioplanktonmetabolism, but also a shift of production fromsmaller, free-living cells to that of particle-associ-ated and/or larger, rapidly growing free-living bac-teria. The shift of bacterioplankton production tothe attached fraction under enriched conditionsmay represent an important emergent property inestuarine systems (CRUMP, BAROSS, and SIMEN-STAD, 1998; CRUMP and BAROSS, 2000) that hasfar-reaching implications with respect to our abil-ity to accurately assess carbon flux in naturalaquatic systems (BIDDANDA, OGDAHL, and COT-NER, 2001; COTNER and BIDDANDA, 2002).

Effect of Freshwater Inputs: Monie Creek vs.Little Monie Creek

Despite consistently elevated nutrient concen-trations in MC, bacterial production in this systemwas consistently lower than that of LMC and onlymarginally higher than that of LC when 2-yearmeans and individual sampling events were con-sidered (Figures 3 and 5, respectively). Small-scaleincubation experiments conducted in the fall of2000 revealed a similar phenomenon (data notshown), namely the lack of a productive responseto inorganic nutrient enrichments by bacterio-plankton from MC. Relatively low rates of BP inMC suggest that there are systematic differencesin environmental conditions between MC andLMC that mediate the effect of nutrient enrich-ment on bacterioplankton metabolism. We hypoth-esize that low-quality terrestrial DOM—as evi-denced by elevated CDOM (Table 1) and d13C sig-natures of terrestrial C3 plants (STRIBLING andCORNWELL, 1997)—drives the muted response tonutrients observed in MC, although the direct ef-fect of salinity on bacterioplankton community me-tabolism and phylogeny may also be important.

Elevated DOC concentrations in MC relative toother creeks (Table 1, Figure 3) would ostensiblysuggest that bacterioplankton production shouldnot be carbon limited in this system (BAINES andPACE, 1991; VALLINO, HOPKINSON, and HOBBIE,1996), and therefore bacterioplankton should befree to respond productively to increases in ambi-ent nutrient concentrations. Significant inputs offresh water to this system (JONES, MURRAY, andCORNWELL, 1997) are accompanied by an increase



in the input of terrestrially derived organic matter,as evidenced by measurements of CDOM (Table 1,Figure 3) and stable isotope analysis (STRIBLING

and CORNWELL, 1997). Although the watershedsof MC and LMC are similar with respect to thepercent of forested land (Figure 1), MC is charac-terized by more extensive forested uplands. Or-ganic matter from terrestrial sources typically haselevated concentrations of high-molecular-weightDOM (MCKNIGHT et al., 2001) that tends to bemore refractory (SUN et al., 1997) and thereforeyields lower growth efficiencies (GOLDMAN, CA-RON, and DENNET, 1987) and lower rates of bac-terial production (MORAN and HODSON, 1990;AMON and BENNER, 1996). It is therefore likelythat BP itself is functionally carbon limited, drivenby lower growth efficiencies and the dominance oflow-quality, terrestrially derived substrates in thissystem.

Bacterial growth efficiency (BGE) is highly var-iable among aquatic systems (DEL GIORGIO andCOLE, 1998, 2000) and on small spatial scaleswithin estuarine systems (DEL GIORGIO and BOU-VIER, 2002). Based on the lack of coherence be-tween BP and bacterial respiration (BR) associat-ed with highly variable BGE, BP alone is a poorpredictor of total carbon flux, and lower bacterialproduction in MC does not necessarily translateinto a similar reduction in total carbon consump-tion. In fact, it is likely that bacterial respirationin MC is high relative to BP, driven by the in-creased metabolic demands of processing and in-corporating refractory organic matter into bacte-rial biomass (LINTON and STEVENSON, 1978) orthe direct effect of changes in dissolved nutrientstoichiometry on BR (CIMBLERIS and KALFF,1998). In addition, shifts in ambient salinities thatoccur during tidal mixing when low-salinity head-waters meet high-salinity water from the bay maystress estuarine bacterioplankton communities,causing mortality and inhibiting growth (DEL

GIORGIO and BOUVIER, 2002). Over a distance ofless than 4 kilometers, bacterioplankton commu-nities in MC may be exposed to a salinity rangefrom ,1‰ in the upper estuary to .13‰ in theOB (data not shown), a much larger range thanthat of LMC and potentially generating a gradientadequate for disrupting bacterioplankton commu-nity metabolism (DEL GIORGIO and BOUVIER,2002), thereby producing lower growth efficienciesand lower rates of production. Because the rate atwhich organic matter is regenerated into dissolvednutrients or is available for consumption by higher

trophic levels is a direct function of BGE (SHERR

and SHERR, 1988; JORGENSEN et al., 1999; KIRCH-MAN, 2000), it is impossible to accurately predictmicrobially mediated changes in carbon flux andnutrient cycling in aquatic systems without inde-pendent assessments of both bacterial productionand respiration.

The effect of substrate quality on BP in fresh-water-dominated MC may be accompanied by thedirect effect of salinity itself on the phylogeneticcomposition of bacterioplankton communities. Re-cent studies have identified dramatic shifts in phy-logenetic composition of natural bacterial assem-blages along salinity gradients, with the domi-nance of specific phylogenetic groups associatedwith certain salinity regimes (CRUMP, ARMBRUST,and BAROSS, 1999; DEL GIORGIO and BOUVIER,2002). In turn, phylogenetic composition of naturalbacterial assemblages has been linked to bacter-ioplankton metabolic properties (PINHASSI et al.,1999; BOUVIER and DEL GIORGIO, 2002) and eventhe use of specific organic substrates (COTTRELL

and KIRCHMAN, 2000). Thus, we predict that dif-ferences in the metabolic response of bacterio-plankton to nutrient enrichment of MC vs. LMCmay be driven by differences in organic matterquality, as well as changes in the phylogeneticcomposition of resident bacterioplankton assem-blages and the unique metabolic capacities asso-ciated with these phylotypes.

Response to Pulsed Nutrient Inputs

We investigated the response of bacterioplank-ton to temporal changes in nutrient enrichment bycomparing estimates of BP during and followingperiods of nutrient input from this watershed.LMC was selected for these comparisons becausethis system has demonstrated a strong response tonutrient enrichment, as evidenced by field obser-vations (Tables 2 and 3, Figures 3, 4, and 5), ma-nipulative nutrient enrichment experiments (un-published data), estimates of chlorophyll a concen-trations, and changes in the marsh macrophytecommunity associated with episodic nutrient deliv-ery (JONES, MURRAY, and CORNWELL, 1997).Pulsed nutrient inputs in July 2000 resulted in el-evated nutrient concentrations (50 and 1 mM forTDN and TDP, respectively) relative to those inSeptember (Figure 5). Similarly, BP was signifi-cantly higher in July relative to September (6.8 vs.2.6 micrograms C liter21 h21, respectively). Thesame comparison of BP in MC during these sum-



mer months revealed a muted response of bacter-ioplankton to nutrient enrichment that has be-come characteristic of this particular creek system(Table 1, Figure 3). Despite higher nutrient con-centrations in July than in September for MC (73vs. 41 mM and 3.0 vs. 0.6 mM for TDN and TDP,respectively), there was not a significant differencein BP (2.2 vs. 2.0 mM; Figure 5). Intermediate con-centrations of TDN and TDP (55 and 1.7 mM, re-spectively) in August may have stimulated themarginally higher BP at this time (3.1 microgramsC liter21 h21), although this was probably an effectof higher temperature (28.08C) on BP (SHIAH andDUCKLOW, 1994). As a result of sampling difficul-ties, estimates of BP in LMC during August werenot available.

During April 2000, the bacterioplankton com-munity in LMC was exposed to elevated TDN andTDP (90 and 4 mM, respectively; Figure 5), deliv-ered as a result of spring fertilizer applicationsand runoff events. The following month, nutrientconcentrations in LMC were much lower and wellbelow the overall mean for this system (Table 1).Although a comparison of BP between April andMay (2.2 and 4.3 micrograms C liter21 h21, respec-tively) does not initially indicate a positive re-sponse to nutrients, SHIAH and DUCKLOW (1994)conducted a series of studies in marshes similar tothose of Monie Bay and found that BP is regulatedpredominantly by temperature in small estuarinesystems during nonsummer months. The authorsreport an average Q10 value of 2.7 (6 0.3) for bac-terial growth in the temperature range of 3–258C.We hypothesized that the bacterioplankton com-munity in April may have been constrained bytemperature and therefore was less responsive tosystem-level nutrient enrichment. This is furthersupported by the general coherence of temperatureand BP when monthly sampling (Figure 5) andcomparison of seasonal means (Figure 6) are con-sidered. Using the reported Q10 value of SHIAH andDUCKLOW (1994) and a difference in ambient wa-ter temperature for April and May of 108C (138Cvs. 238C, respectively; Figure 5), we calculatedtemperature-corrected estimates of BP for these 2months and estimated that BP in April would havebeen elevated relative to that of May (6.0 vs. 4.3micrograms C liter21 h21, respectively). Based onthe comparisons of BP during summer months andtemperature-corrected estimates of BP in Apriland May, we conclude that, although bacterio-plankton respond positively to increases in ambi-ent nutrient concentrations, temperature is an im-

portant environmental factor mediating the mag-nitude of this response and should be consideredin seasonal comparisons of BP in temperate sys-tems.

CONCLUSIONS

We observed a persistent response of bacterio-plankton to agriculturally driven enrichment ofthe tidal creeks, a conclusion that correspondswith generally held paradigms regarding the effectof nutrients on natural heterotrophic bacterio-plankton communities (KIRCHMAN, 2000). How-ever, not all systems responded to nutrient enrich-ment in a similar manner, and freshwater inputsand/or salinity play an important role in mediatingthe effect of nutrients on estuarine bacterioplank-ton communities. We attribute the muted responseto enrichment observed at lower salinities to anabundance of refractory, terrestrially derived or-ganic matter and/or the direct effect of salinity onbacterioplankton, which in turn drives changes insubstrate quality, assemblage phylogenetic com-position, and ultimately bacterioplankton com-munity metabolic processes.

The metabolic response of bacterioplankton tonutrient enrichment is extremely complex, occur-ring at both the cellular (DEL GIORGIO and SCAR-BOROUGH, 1995; CHOI, SHERR, and SHERR, 1999)and community (VREDE et al., 1999; PACE andCOLE, 2000) levels. As a result, we recognize thatestimates of BP and BA alone cannot accuratelycapture subtle changes in bacterioplankton metab-olism and together are inadequate to identify un-equivocally mechanisms underlying the disparateresponse of MC and LMC to nutrient enrichment.When coupled with BP and BA, estimates of sin-gle-cell activity such as DNA content (GASOL andDEL GIORGIO, 2000; LEBARON et al., 2001a) or theabundance of actively respiring cells (RODRIGUEZ

et al., 1992) may provide a more sensitive index ofbacterioplankton metabolism. Similarly, combin-ing estimates of BP and BR not only yields an es-timate of bacterial growth efficiency, but also ameasure of the total carbon consumed by the bac-terioplankton community. A comprehensive suiteof cellular and community-level indices of bacter-ioplankton metabolism is essential for accurate as-sessment of microbially mediated carbon and nu-trient cycling in aquatic systems, and the investi-gation of these parameters in Monie Bay may ul-timately lead to important insight regardingmechanisms underlying the response of bacterio-



plankton to system-level nutrient enrichment ofestuarine systems.

The effect of system-level enrichment on estua-rine systems is reflected in numerous aspects ofmarsh ecology (CLOERN, 2001), including phyto-plankton abundance (ANDERSON and TAYLOR,2001), macrophyte community diversity and pro-duction (DAY et al., 1989; VALIELA, 1995), and sed-iment biogeochemical processes (CORNWELL et al.,1996; DAUER, WEISBERG, and RANASINGHE,2000). The metabolic response of bacterioplanktonto system-level nutrient enrichment is not as welldocumented (HOPPE, GIESENHAGEM, and GOCKE,1998; LEBARON et al., 1999), although it may rep-resent a more sensitive and integrative assess-ment than other traditional indices of ecosystemfunction and eutrophication. The bacterioplanktoncommunity responds rapidly (i.e., hours to days) tochanges in environmental conditions and is wellsuited for investigations of tidally influenced sys-tems where episodic and pulsed nutrient inputsare common. However, the persistence of system-specific patterns in bacterial production among thecreeks of Monie Bay during our 2-year study pe-riod also suggests an integration of conditions overmuch longer time periods. Combining multiple as-pects of bacterioplankton metabolism in investi-gations of estuarine systems may provide an ex-tremely comprehensive and multifaceted index ofecosystem function that reflects changes in sys-tem-level processes on multiple spatial and tem-poral scales.

ACKNOWLEDGMENTS

We thank T.W. Jones for the generous contri-bution of background data for Monie Bay, Chesa-peake Bay Maryland NERR Research CoordinatorJulie Bortz for facilitating our participation in thisspecial issue, Kitty Fielding for extensive assis-tance with fieldwork, and the panel of anonymousreviewers for their time, effort, and valuable con-tributions to this article.

LITERATURE CITED

AMON, R. and BENNER, R., 1996. Bacterial utilization ofdifferent size classes of dissolved organic matter. Lim-nology and Oceanography, 41, 41–51.

ANDERSON, J.; HARDY, E.; ROACH, J., and WITMER, R.,1976. A land use and land cover classification systemfor use with remote sensor data. Washington, D.C.:United States Department of the Interior.

ANDERSON, T.H. and TAYLOR, G.T., 2001. Nutrient puls-es, plankton blooms, and seasonal hypoxia in westernLong Island Sound. Estuaries, 24, 228–243.

AZAM, F.; FENCHEL, T.; FIELD, J.G.; GRAY, J.S.; MEYER-REIL, L.A., and THINGSTAD, T.F., 1983. The ecologicalrole of water-column microbes in the sea. Marine Ecol-ogy Progress Series, 10, 257–263.

BAINES, S.B. and PACE, M.L., 1991. The production ofdissolved organic matter by phytoplankton and its im-portance to bacteria: patterns across marine andfreshwater systems. Limnology and Oceanography, 36,1078–1090.

BANO, N.; NISA, M.; KHAN, N.; SALEEM, M.; HARRISON,P.J.; AHMED, S., and AZAM, F., 1997. Significance ofbacteria in the flux of organic matter in the tidalcreeks of the mangrove ecosystem of the Indus Riverdelta, Pakistan. Marine Ecology Progress Series, 157,1–12.

BEAULAC, M.N. and RECKHOW, K.H., 1982. An exami-nation of land use—nutrient export relationships. Wa-ter Resources Bulletin, 18, 1013–1024.

BIDDANDA, B.; OGDAHL, M., and COTNER, J.B., 2001.Dominance of bacterial metabolism in oligotrophic rel-ative to eutrophic waters. Limnology and Oceanogra-phy, 46, 730–738.

BOUVIER, T.C. and DEL GIORGIO, P.A., 2002. Composi-tional changes in free-living bacterial communitiesalong a salinity gradient in two temperate estuaries.Limnology and Oceanography, 47, 453–470.

BRUSSAARD, C. and RIEGMAN, R., 1998. Influence of bac-teria on phytoplankton cell mortality with phosphorusor nitrogen as the algal-growth-limiting nutrient.Aquatic Microbial Ecology, 14, 271–280.

CARLSON, C. and DUCKLOW, H.W., 1996. Growth of bac-terioplankton and consumption of dissolved organiccarbon in the Sargasso Sea. Aquatic Microbial Ecology,10, 69–85.

CARON, D.A.; LIM, E.L.; SANDERS, R.W.; DENNETT,M.R., and BERNINGER, U.G., 2000. Responses of bac-terioplankton and phytoplankton to organic carbonand inorganic nutrient additions in contrasting oce-anic ecosystems. Aquatic Microbial Ecology, 22, 175–184.

CHOI, J.; SHERR, E.B., and SHERR, B.F., 1999. Dead oralive? A large fraction of ETS-inactive marine bacter-ioplankton cells, as assessed by reduction of CTC, canbecome ETS-active with incubation and substrate ad-dition. Aquatic Microbial Ecology, 18, 105–115.

CIMBLERIS, A.C. and KALFF, J., 1998. Planktonic bac-terial respiration as a function of C:N:P ratios acrosstemperate lakes. Hydrobiologia, 384, 89–100.

CLOERN, J., 2001. Our evolving conceptual model of thecoastal eutrophication problem. Marine Ecology Pro-gress Series, 210, 223–253.

COLE, J.J.; FINDLAY, S., and PACE, M.L., 1988. Bacterialproduction in fresh and saltwater ecosystems: a crosssystem overview. Marine Ecology Progress Series, 43,1–10.

CORNWELL, J.C.; CONLEY, D.J.; OWENS, M., and STE-VENSON, J.C., 1996. A sediment chronology of the eu-trophication of Chesapeake Bay. Estuaries, 19, 488–499.

CORNWELL, J.C.; STRIBLING, J.M., and STEVENSON,J.C., 1994. Biogeochemical Studies at the Monie BayNational Estuarine Research Reserve. Washington,D.C.: Organizing for the Coast: Thirteenth Interna-tional Conference of the Coastal Society, pp. 645–655.



COTNER, J.B. and BIDDANDA, B., 2002. Small players,large role: microbial influence on biogeochemical pro-cesses in pelagic aquatic ecosystems. Ecosystems, 5,105–121.

COTTRELL, M.T. and KIRCHMAN, D.L., 2000. Natural as-semblages of marine proteobacteria and members ofCytophaga-Flavobacter cluster consuming low- andhigh-molecular weight dissolved organic matter. Ap-plied and Environmental Microbiology, 66, 1692–1697.

CRUMP, B.; ARMBRUST, E., and BAROSS, J., 1999. Phy-logenetic analysis of particle attached and free-livingbacterial communities in the Columbia River, its es-tuary, and the adjacent coastal ocean. Applied and En-vironmental Microbiology, 65, 3192–3204.

CRUMP, B.C.; BAROSS, J.A., and SIMENSTAD, C.A., 1998.Dominance of particle-attached bacteria in the Colum-bia River estuary, USA. Aquatic Microbial Ecology, 14,7–18.

CRUMP, B.C. and BAROSS, J.A., 2000. Characterizationof the bacterially-active particle fraction in the Colum-bia River estuary. Marine Ecology Progress Series,206, 13–22.

DAUER, D.; WEISBERG, S., and RANASINGHE, J., 2000.Relationships between benthic community condition,water quality, sediment quality, nutrient loads, andland use patterns in Chesapeake Bay. Estuaries, 23,80–96.

DAY, J.W.J.; HALL, C.A.S.; KEMP, W.M., and YANEZ-AR-ANCIBIA, A., 1989. Estuarine Ecology. New York: JohnWiley and Sons.

DEL GIORGIO, P.A.; BIRD, D.F.; PRAIRIE, Y.T., and PLAN-AS, D., 1996a. Flow cytometric determination of bac-terial abundance in lake plankton using the green nu-cleic acid stain SYTO 13. Limnology and Oceanogra-phy, 41, 783–789.

DEL GIORGIO, P.A. and BOUVIER, T.C., 2002. Linkingthe physiologic and phylogenetic successions in free-living bacterial communities along an estuarine salin-ity gradient. Limnology and Oceanography, 47, 471–486.

DEL GIORGIO, P.A. and COLE, J.J., 1998. Bacterialgrowth efficiency in natural aquatic systems. AnnualReview of Ecological Systematics, 29, 503–541.

DEL GIORGIO, P.A. and COLE, J.J., 2000. Bacterialgrowth energetics and efficiency in natural aquaticsystems. In: KIRCHMAN, D.L. (ed.), Microbial Ecologyof the Oceans. New York: John Wiley and Sons, pp.289–325.

DEL GIORGIO, P.A.; GASOL, J.M.; VAQUE, D.; MORA, P.;AGUSTI, S., and DUARTE, C.M., 1996b. Bacterioplank-ton community structure: protists control net produc-tion and the proportion of active bacteria in a coastalmarine community. Limnology and Oceanography, 41,1169–1179.

DEL GIORGIO, P.A. and SCARBOROUGH, G., 1995. In-crease in the proportion of metabolically active bac-teria along gradients of enrichment in freshwater andmarine plankton: implications for estimates of bacte-rial growth and production rates. Journal of PlanktonResearch, 17, 1905–1924.

DUCKLOW, H.W.; PURDIE, D.A.; WILLIAMS, P.J.L.B., andDAVIES, J.M., 1986. Bacterioplankton: a sink for car-bon in a coastal marine plankton community. Science,232, 863–867.

FIELDING, K.P., 2003. Differential Substrate Limitationin Small Tributaries of Chesapeake Bay, Maryland In-fluenced by Non-point Source Nutrient Loading. Col-lege Park: University of Maryland, Master’s thesis.

FINLAY, B.J.; MABERLY, S.C., and COOPER, J.I., 1997.Microbial diversity and ecosystem function. Oikos, 80,209–213.

FISHER, T.R., 1985. Nitrogen and phosphorus inputs toChesapeake Bay. Williamsburg, VA: The ChesapeakeBay Symposium, National Marine Educators Confer-ence, pp. 58–62.

GASOL, J.M. and DEL GIORGIO, P.A., 2000. Using flowcytometry for counting natural planktonic bacteriaand understanding the structure of planktonic bacte-rial communities. Scientia Marina, 64, 197–224.

GOLDMAN, J.C.; CARON, D.A., and DENNET, M.R., 1987.Regulation of gross growth efficiency and ammoniumregeneration in bacteria by C:N ratio. Limnology andOceanography, 32, 1239–1252.

GONZALEZ, J.M.; SHERR, E.B., and SHERR, B.F., 1990.Size-selective grazing on bacteria by natural assem-blages of estuarine flagellates and ciliates. Appliedand Environmental Microbiology, 56, 583–589.

GOOSEN, N.K.; VAN RIJSWIJK, P.; KROMKAMP, J., andPEENE, J., 1997. Regulation of annual variation inheterotrophic bacterial production in the Schelde es-tuary (SW Netherlands). Aquatic Microbial Ecology,12, 223–232.

HOCH, M.P. and KIRCHMAN, D.L., 1993. Seasonal andinter-annual variability in bacterial production andbiomass in a temperate estuary. Marine Ecology Pro-gress Series, 98, 283–295.

HOPPE, H.G.; GIESENHAGEN, H.C., and GOCKE, K.,1998. Changing patterns of bacterial substrate decom-position in a eutrophication gradient. Aquatic Micro-bial Ecology, 15, 1–13.

HU, C.; MULLER-KARGER, F.E., and ZEPP, R.G., 2002.Absorbance, absorption coefficient, and apparentquantum yield: a comment on common ambiguity inthe use of these optical concepts. Limnology andOceanography, 47, 1261–1267.

JONES, T.W.; MURRAY, L., and CORNWELL, J.C., 1997.A two-year study of the short-term and long-term se-questering of nitrogen and phosphorus in the Mary-land National Estuarine Research Reserve. MonieBay, MD: Maryland National Estuarine Research Re-serve, pp. 1–100.

JORGENSEN, N.; KROER, N.; COFFIN, R.B., and HOCH,M.P., 1999. Relations between bacterial nitrogen me-tabolism and growth efficiency in an estuarine and anopen-water ecosystem. Aquatic Microbial Ecology, 18,247–261.

KEARNEY, M.S.; STEVENSON, J.C., and WARD, L.G.,1994. Spatial and temporal changes in marsh verticalaccretion rates at Monie Bay—implications for sea-level rise. Journal of Coastal Research, 10, 1010–1020.

KIRCHMAN, D.L., 1993. Leucine incorporation as a mea-sure of biomass production by heterotrophic bacteria.In: KEMP, P.F.; SHERR, B.F.; SHERR, E.B., and COLE,J.J. (eds.), Handbook of Methods in Aquatic MicrobialEcology. Boca Raton, FL: CRC Press, pp. 776.

KIRCHMAN, D.L., 1994. The uptake of inorganic nutri-ents by heterotrophic bacteria. Microbial Ecology, 28,255–271.



KIRCHMAN, D.L., 2000. Microbial Ecology of the Oceans.New York: Wiley-Liss.

KOLBER, Z.S.; BARBER, R.; COALE, K.; FITZWATER, S.,and GREENE, R., 1994. Iron limitation of phytoplank-ton photosynthesis in the equatorial Pacific Ocean.Nature, 371, 145–148.

LEBARON, P.; SERVAIS, P.; AGOGUE, H.; COURTIES, C.,and JOUX, F., 2001a. Does the high nucleic acid con-tent of individual bacterial cells allow us to discrimi-nate between active cells and inactive cells in aquaticsystems? Applied and Environmental Microbiology, 67,1775–1782.

LEBARON, P.; SERVAIS, P.; TROUSSELLIER, M.; COUR-TIES, C.; MUYZER, G.; BERNARD, L.; SCHAEFER, H.;PUKALL, R.; STACKEBRANDT, E.; GUINDULAIN, T., andVIVES-REGO, J., 2001b. Microbial community dynam-ics in Mediterranean nutrient-enriched seawater me-socosms: changes in abundances, activity and compo-sition. FEMS Microbiology Ecology, 34, 255–266.

LEBARON, P.; SERVAIS, P.; TROUSSELLIER, M.; COUR-TIES, C., and VIVES-REGO, J., 1999. Changes in bac-terial community structure in seawater mesocosmsdiffering in their nutrient status. Aquatic MicrobialEcology, 19, 225–267.

LEE, K.Y.; FISHER, T.R.; JORDAN, T.E.; CORREL, D.L.,and WELLER, D.E., 2000. Modeling the hydrochemis-try of the Choptank River basin using GWLF and arc/info: 1. Model calibration and validation. Biogeochem-istry, 49, 143–173.

LINTON, J. and STEVENSON, R., 1978. A preliminarystudy on growth yields in relation to the carbon andenergy content of various organic growth substrates.FEMS Microbiology Letters, 3, 95–98.

LOWRANCE, R.; ALTIER, L.S.; NEWBOLD, J.D.; SCHNA-BEL, R.R.; GROFFMAN, P.M.; DENVER, J.M.; CORRELL,D.L.; GILLIAM, J.W.; ROBINSON, J.L.; BRINSFIELD,R.B.; STAVER, K.W.; LUCAS, W., and TODD, A.H.,1997. Water quality functions of riparian forest buff-ers in Chesapeake Bay watersheds. EnvironmentalManagement, 21, 687–712.

MCKNIGHT, D.; BOYER, E.; WESTERHOFF, P.; DORAN, P.;KULBE, T., and ANDERSON, D., 2001. Spectrofluoro-metric characterization of dissolved organic matter forindication of precursor organic material and aromatic-ity. Limnology and Oceanography, 46, 38–48.

MORAN, M.A. and HODSON, R.E., 1990. Bacterial pro-duction on humic and non-humic components of dis-solved organic matter. Limnology and Oceanography,35, 1744–1756.

MORAN, M.A.; SHELDON, J.E., and ZEPP, R.G., 2000.Carbon loss and optical property changes during long-term photochemical biological degradation of estua-rine dissolved organic matter. Limnology and Ocean-ography, 45, 1254–1264.

NIXON, S.W., 1995. Coastal marine eutrophication: a def-inition, social causes, and future concerns. Ophelia,41, 199–219.

NORTON, M.M. and FISHER, T.R., 2000. The effects offorest on stream water quality in two coastal plain wa-tersheds of the Chesapeake Bay. Ecological Engineer-ing, 14, 337–362.

PACE, M.L. and COLE, J.J., 2000. Effects of whole-lakemanipulations of nutrient loading and food web struc-

ture on planktonic respiration. Canadian Journal ofFisheries and Aquatic Science, 57, 487–496.

PAINCHAUD, J.; LEFAIVRE, D.; THERRIAULT, J.C., andLEGENDRE, L., 1996. Bacterial dynamics in the upperSt. Lawrence estuary. Limnology and Oceanography,41, 1610–1618.

PINHASSI, J.; AZAM, F.; HEMPHALA, J.; LONG, R.; MAR-TINEZ, J.; ZWEIFEL, U., and HAGSTROM, A., 1999. Cou-pling between bacterioplankton species composition,population dynamics, and organic matter degradation.Aquatic Microbial Ecology, 17, 13–26.

RASMUSSEN, B. and JOSEFSON, A., 2002. Consistent es-timates for the residence time of micro-tidal estuaries.Estuarine, Coastal and Shelf Science, 54, 65–73.

REITNER, B.; HERZIG, A., and HERNDL, G., 1999. Dy-namics in bacterioplankton production in a shallow,temperate lake (Lake Neusiedl, Austria): Evidence fordependence on macrophyte production rather than onphytoplankton. Aquatic Microbial Ecology, 19, 245–254.

REVILLA, M.; IRIARTA, A., and ORIVE, E., 2000. Bacterialand phytoplankton dynamics along a trophic gradientin a shallow temperate estuary. Estuarine, Coastaland Shelf Science, 50, 297–313.

RIEMAN, B.; SORENSEN, H.M.; BJORNSEN, P.K.; HORST-ED, S.J.; JENSEN, L.M.; NIELSEN, T.G., and SONDER-GAARD, M., 1990. Carbon budgets of the microbial foodweb in estuarine enclosures. Marine Ecology ProgressSeries, 65, 159–170.

RODRIGUEZ, G.G.; PHIPPS, D.; ISHIGURO, K., and RIDG-WAY, H.F., 1992. Use of a fluorescent redox probe fordirect visualization of actively respiring bacteria. Ap-plied and Environmental Microbiology, 58, 1801–1808.

SCUDLARK, J.R. and CHURCH, T.M., 1989. The sedimen-tary flux of nutrients at a Delaware salt-marsh site—a geochemical perspective. Biogeochemistry, 7, 55–75.

SHARP, J.; BENNER, R.; BENNET, L.; CARLSON, C.; FITZ-WATER, S.; PELTZER, E., and TUPAS, L., 1995. Analy-ses of dissolved organic carbon in sea water: theJGOFS EqPac methods comparison. Marine Chemis-try, 48, 91–108.

SHERR, E.B. and SHERR, B.F., 1988. Role of microbes inpelagic food webs: a revised concept. Limnology andOceanography, 33, 1225–1227.

SHIAH, F.K. and DUCKLOW, H.W., 1994. Temperatureand substrate regulation of bacterial abundance, pro-duction and specific growth rate in Chesapeake Bay,USA. Marine Ecology Progress Series, 103, 297–308.

SHIAH, F.K. and DUCKLOW, H.W., 1995. Multiscale var-iability in bacterioplankton abundance, production,and specific growth rate in a temperate salt-marshtidal creek. Limnology and Oceanography, 40, 55–66.

SIMS, J.T. and WOLF, D.C., 1994. Poultry waste man-agement: agricultural and environmental issues. Ad-vances in Agronomy, 52, 1–83.

SIMS, J.T.; SIMARD, R.R., and JOERN, B.C., 1998. Phos-phorus loss in agricultural drainage: historical per-spective and current research. Journal of Environmen-tal Quality, 27, 277–293.

SMITH, D.C. and AZAM, F., 1992. A simple, economicalmethod for measuring bacterial protein synthesisrates in seawater using super(3)H-leucine. Marine Mi-crobial Food Webs, 6, 107–114.

SMITH, D.E.; LEFFLER, M., and MACKIERNAN, G., 1992.



Oxygen Dynamics in the Chesapeake Bay: A Synthesisof Recent Research. College Park, MD: Maryland SeaGrant.

SMITH, E. and KEMP, W., 2001. Size structure and theproduction/respiration balance in a coastal planktoncommunity. Limnology and Oceanography, 46, 473–485.

SPEIRAN, G.K.; HAMILTON, P.A., and WOODSIDE, M.D.,1998. Natural processes for managing nitrate inground water discharged to the Chesapeake Bay andother surface waters: more than forest buffers. Rich-mond, VA: USGS, pp. 6.

STAVER, K.W. and BRINSFIELD, R.B., 2001. Agricultureand water quality on the Maryland eastern shore:where do we go from here? BioScience, 51, 859–868.

STRIBLING, J.M. and CORNWELL, J.C., 1997. Identifica-tion of important primary producers in a ChesapeakeBay tidal creek system using stable isotopes of carbonand sulfur. Estuaries, 20, 77–85.

STRIBLING, J.M. and CORNWELL, J.C., 2001. Nitrogen,phosphorus, and sulfur dynamics in a low salinitymarsh system dominated by Spartina alterniflora.Wetlands, 21, 629–638.

STRICKLAND, J. and PARSONS, T., 1972. A PracticalHandbook of Seawater Analysis. Bulletin of the Fish-eries Research Board of Canada, 167, 1–310.

SUN, L.; PERDUE, E.; MEYER, J., and WEIS, J., 1997. Useof elemental composition to predict bioavailability ofdissolved organic matter in a Georgia river. Limnologyand Oceanography, 42, 714–721.

TUTTLE, J.; JONAS, R., and MALONE, T., 1987. Origin,

development and significance of Chesapeake Bay an-oxia. Conference 152. Philadelphia, PA: NationalMeeting AAAS, Chesapeake Bay Fisheries and Con-taminant Problems, pp. 442–472.

VALDERRAMA, J.C., 1981. The simultaneous analysis oftotal nitrogen and total phosphorus in natural waters.Marine Chemistry, 10, 109–122.

VALIELA, I., 1995. Marine Ecological Processes. NewYork: Springer-Verlag.

VALLINO, J.J.; HOPKINSON, C.S., and HOBBIE, J.E.,1996. Modeling bacterial utilization of dissolved or-ganic matter: optimization replaces Monod growth ki-netics. Limnology and Oceanography, 41, 1591–1609.

VREDE, K.; VREDE, T.; TISAKSSON, A., and KARLSSON,A., 1999. Effects of nutrients (P, N, C) and zooplank-ton on bacterioplankton and phytoplankton—a sea-sonal study. Limnology and Oceanography, 44, 1616–1624.

WARD, L.G.; KEARNEY, M.S., and STEVENSON, J.C.,1998. Variations in sedimentary environments and ac-cretionary patterns in estuarine marshes undergoingrapid submergence, Chesapeake Bay. Marine Geology,151, 111–134.

WEIL, R.R.; WEISMILLER, R.A., and TURNER, R.S., 1990.Nitrate contamination of groundwater under irrigatedcoastal plain soils. Journal of Environmental Quality,19, 441–448.

WHITLEDGE, T.C.; MALLORY, S.C.; PATTON, C.J., andWIRICK, C.D., 1981. Automated nutrient analysis inseawater. Upton, NY: Department of Energy and En-vironment.

The Effects of System-Level Nutrient Enrichment on...

Documents

Transcript of The Effects of System-Level Nutrient Enrichment on...