Fluxes of bacterioplankton between a tidal estuary and the sea: returningto the “Outwelling Hypothesis”

M.A. Cunha1,*, J.M. Dias2, M.A. Almeida1, J.F. Lopes2 and F. Alcântara1

1Department of Biology, University of Aveiro, Campo Universitário de Santiago, Aveiro, 3810-193, Portugal;2Department of Physics, University of Aveiro, Campo Universitário de Santiago, Aveiro, 3810-193, Portugal;*Author for correspondence (e-mail: acunha@bio.ua.pt)

Received 18 December 2001; accepted in revised form 2 September 2002

Key words: Estuarine bacterioplankton, Tidal fluxes

Abstract

The tidal dynamics of bacterioplankton communities at the outer part of a shallow estuary (Ria de Aveiro, Por-tugal) were studied during 6 tidal cycles at a fixed sampling site. Bacterial numbers (0.2–8.1 × 109 cells l−1),aminopeptidase activity (189–1662 nmol l−1 h−1), �-glucosidase activity (1.7–67.0 nmol l−1 h−1) and potentialglucose incorporation (0.48–3.99 nmol l−1 h−1) followed a consistent pattern of increase during ebb and decreaseduring flood. Fluxes of bacterioplankton populations and associated heterotrophic activities between the estuaryand the coastal area during a tidal cycle were estimated from the water flux as derived from a two-dimensionalvertically-integrated numerical model. The net fluxes estimated for a tidal cycle ranged from −26.0 to −2.5 to ×1016 bacterial cells. The net tidal fluxes of potential heterotrophic activities ranged from −10 to −80 mol h−1 foraminopeptidase, −0.33 to −1.10 mol h−1 for �-glucosidase and −0.18 to +0.03 mol h−1 for glucose incorporation.Net fluxes were generally negative in sign indicating the transfer of phyto- and bacterioplankton, as well as po-tential capacities for the degradation and recycling of organic matter, from the outer estuarine compartment tothe sea.

Introduction

Estuaries are interface ecosystems that functionallycouple continental and marine environments, receiv-ing biogeochemically active inputs from land, riversand the coastal sea. Estimates of the world-wide in-put of sediments into the ocean by rivers, coastal ero-sion, wind, etc., vary from 8 to 64 × 109 tons per year(Milliman and Meade 1983). It was also estimatedthat the annual transport of carbon from rivers to theocean corresponds to 215 × 106 tons of dissolved or-ganic carbon and 180 × 106 tons of particulate or-ganic carbon (Martin and Gordeev 1984) and thefluxes of mineral nutrients, such as phosphate and ni-trate to the oceans, have world-wide more than dou-bled in the last decades (Meybeck 1998).

The estuarine environment is often reported as ex-tremely productive, supporting dense bacterioplank-ton communities and high rates of primary production

and bacterial heterotrophic activity (Barbosa 1991;Fuks et al. 1991; Hoppe et al. 1996; Revilla et al.2000). The basic model of salt-marsh estuaries as ex-porting systems usually referred to as the “OutwellingHypothesis” (Odum 1968; Dame et al. 1986) devel-oped towards the notion that marsh productivity maybe “outwelled” as organisms rather than as organicmatter and nutrients (Odum 2000).

Major transformations of the pools of organic andinorganic nutrients that reach the main body of theestuary are related to biological processes associatedwith bacterioplankton (Billen and Garnier 1997; Clo-ern 2001). This, in turn, strongly responds to environ-mental stimuli (Cunha et al. 2001), namely variationsin nutrient availability (Hoppe et al. 1998). Highfresh-water inputs and the corresponding increase inthe concentration of labile organic compounds en-hance the levels of heterotrophic activity of estuarinebacteria (Murrell et al. 1999) and impose a river-

45Aquatic Ecology 37: 45–54, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

driven seasonality in the way the estuary interactswith the nearby ocean (Cloern and Nichols 1985).However, in estuaries where the processes of mate-rial transport are mostly driven by tidal fluxes, thesemi-diurnal variability will prevail over seasonal ef-fects (Smith 2001). The tidal flooding of salt-marshesand mud-flats in addition to the advection of dis-solved nutrients from the sediments stimulate bacte-rial activity in the water column (Trousselier et al.1993; von Bodungen et al. 1995; Ritzrau et al. 1997)and probably are major factors for the persistence ofdense and active bacterioplankton communities inshallow estuarine areas (Hoppe et al. 1996; Cunha etal. 2000).

This paper examines the hypothesis that in a shal-low mesotidal estuarine system, such as Ria deAveiro, exports of bacterioplankton and potential het-erotrophic activity to the coastal ocean can occur un-der the summer regime of low fresh water discharge.Tidal dynamics of bacterioplankton were character-ised in terms of (a) patterns of variability of abun-dance and activity and (b) sign and magnitude of thefluxes of cells and potential activities at the interfacewith the sea.

Materials and methods

Study area

Ria de Aveiro (Figure 1) is a bar-built estuary (Prit-chard 1989), also described as a coastal lagoon, onthe Northwest coast of Portugal. It is a complex sys-tem characterised by narrow channels and extensiveintertidal zones. During spring tides, the maximal wetarea is 83 km2 at high tide, reduced to 66 km2 at lowtide (Dias et al. 2000). The estuary encompasses a to-tal volume of water of approximately 160 × 106 m3

at high tide (Silva 1994) and the total river dischargeduring the period equivalent to a tidal cycle is 1.8 ×106 m3 (Moreira et al. 1993). The tidal range variesbetween 0.6 m at neap tides and 3.2 m at spring tides,with an average of about 2 m (Dias et al. 2000).

Study site

The investigation was conducted inside the lagoonassuming that a sampling station located as close aspossible to the mouth would fairly represent the prop-erties of the water masses moving in and out the es-tuary. Station N1 (40° 33� N; 8a 45� W) was located

Figure 1. Ria de Aveiro (Portugal) with sampling estuarine stationN1 indicated with arrow.

46

in the main navigation channel, approximately 2 kmfrom the mouth (Figure 1). Sampling was performedduring day-time during 5 spring tide cycles (01-06-96, 03-06-96, 28-08-96, 30-08-96 and 02-09-97) andone neap tide cycle (10-09-97). In each tidal cycle,water samples were collected just below surface (0.2m) with an horizontal Niskin bottle, at approximately2-hour intervals including samples at low tide slack(LT) and high tide slack (HT), for no longer than 14hours. Samples were transported to the laboratory indark isothermal boxes and were processed within 4hours after collection.

Physical and chemical parameters

Temperature and salinity were measured with a WTWLF 196 Conductivity Meter and total depth of thewater column was determined with a Sonar probe(Hondex PS-7 LCD Digital Sounder).

The concentration of suspended solids (seston) inestuarine samples was determined after filtration of500 ml aliquots through Whatman GF/C (47 mm di-ameter) pre-weighed, pre-combusted filters. The fil-ters were dried at 60 °C for 24 hours and seston con-tent was calculated as the increase in weight.Particulate organic matter (POM) was determined asthe weight loss of the dry seston filters after 4 hoursof incineration at 525 °C (Parsons et al. 1989).

Chlorophyll a was quantified fluorimetrically(Yentsch and Menzel 1963) in a Jasco FP-777 espe-ctrofluorometer. A total volume of 0.5 l of water wasfiltered through Whatman GF/C filters and immedi-ately extracted with 10 ml of 90% acetone. 3 repli-cate determinations were performed on each sample.

Microbiological parameters

Total bacterial number (TBN) was determined by cellcounting under epifluorescence microscopy after fix-ation of water samples with 2% formaldehyde (finalconcentration). The samples (3 replicates) were thenfiltered through 0.2 �m black polycarbonate mem-branes (Poretics) and stained with 0.03% acridine or-ange (Hobbie et al. 1977).

The determination of maximum uptake velocity ofglucose (glucose Vm) followed the procedure de-scribed by Gocke (1977) in 3 replicates and 1 blankof each sample. A final saturation concentration of430 nM of uniformly labelled 14C-glucose (Amer-sham SA 11.5 GBq mmol−1, 310 mCi mmol−1) wasadded to 10 ml aliquots. The saturating concentration

was chosen after kinetic analysis. Incubations werecarried out for 2–3 hours at in situ temperature, in thedark, without agitation. Cells were collected on 0.2�m Poretics polycarbonate membranes and radioac-tivity was read in a liquid scintillation counter (Beck-man LS 6000 IC) using UniverSol as scintillationcocktail.

Ectoenzymatic activity was determined fluorimet-rically (Jasco FP-777 Fluorometer) as the maximumhydrolysis rate (Hm) of model substrates (Sigma-Al-drich) for �-glucosidase (4-methylumbelliferyl-�-D-glucoside) and Leu-aminopeptidase (L-leucine-7-amido-4-methyl-coumarin) according to Hoppe(1983). Saturation was achieved with 1 mM of bothsubstrates. Wavelengths for excitation and emissionwere respectively 380 and 440 nm for MCA (7-ami-no-4-methylcoumarine) and 360 and 450 nm forMUF (4-methylumbelliferone). Measurements weremade on 3 replicates for each sample after 1–2 hoursincubations at in situ temperature. Calibration wasperformed by adding a series of 8 concentrations ofthe fluorescent products (0–500 nM for MUF and 0–6�M for MCA) to a pool of water from all the samplesanalysed each day.

Estimation of estuarine fluxes

The tidal balance of several water properties was es-timated as the difference between the integrated ebband flood tide flows through a defined transverse sec-tion of the channel at station N1. In order to deter-mine estuarine fluxes in tidally induced current fields,water flow and properties concentrations are usuallymeasured in cross-sectional structures along the tidalcycle (Taylor and Allanson 1995). Ria de Aveiro isvertically well mixed and laterally homogeneous(Dias et al. 1999), and therefore it can be consideredthat values measured 0.2 m below surface, and at ap-proximately 2-hours intervals fairly represent thestructure of the correspondent property. In addition,Ria de Aveiro is a very shallow environment whichmakes the use of vertically integrated numerical mod-els adequate to determine the integrated water flow.

In this work a two-dimensional vertically-inte-grated numerical model (2DV) was applied to solvethe well-known shallow-water equations. This modelwas previously calibrated (Dias et al. 1998) and hasbeen used in studies of tidal dynamics of the Ria deAveiro (Dias et al. 2000), where it reproduces thetidal prism values for the main channels of Ria deAveiro computed by other authors (Dias 2001).

47

The 2DV numerical model was used to computetime series of water flow (time-step of 40 s) throughthe transverse section of the channel including stationN1, during each one of the tidal cycles. The times ofslack water were also determined from the numericalresults, allowing the estimation of the time of zeroflux. Instantaneous cross-sectional fluxes of salinity,chlorophyll, POM, ectoenzymatic activities, glucoseincorporation and bacterial abundance were estimatedas the product of the field data measured at a giventime and the simultaneous numerically computed wa-ter flow. New values of salinity, chlorophyll, POM,ectoenzymatic activities, glucose incorporation andbacterial abundance corresponding to the time ofslack water (zero instantaneous fluxes) were also in-cluded.

Both water flow and properties concentrations areknown to have nearly sinusoidal dynamics in Ria deAveiro (Hoppe et al. 1996; Cunha et al. 2000). Timeseries of instantaneous fluxes during a tidal cyclewere produced by polynomial interpolation of thevalue of the products of the field data and the numer-ical results, using a third degree polynomial approxi-mation. As expected, these curves fit very well with

the numerically computed water flow curves. Estua-rine fluxes through the transverse section were latercalculated integrating the ebb and flood tide series.

Results

Tidal fluctuation of water properties andbacterioplankton abundance and activity

Data on the variation of total water depth, tempera-ture and salinity registered during 6 tidal cycles atstation N1 are presented in Table 1. Depth at the sta-tion varied from a minimum of 4.6 m at low tide to10 m at high tide. Water temperature (15.3–20.6 °C)generally decreased during flood and increased dur-ing ebb. Salinity varied within 23.7–35.9 PSU. Salin-ity maxima and minima were always observed at highand low tide, respectively.

Seston (17–222 mg l−1) and POM (6–31 mg l−1)strongly fluctuated during the tidal cycles and a clearpattern could not be identified (Figure 2). The con-centration of chlorophyll a ranged from 6.8 to 10.2�g l−1 with maximal values at low tide (Figure 2).

Table 1. Values of total depth, temperature and salinity at station N1 determined during tidal cycles involving 6 sampling moments: LT (lowtide), LT+2 (2 hours after low tide), HT−2 (2 hours before high tide), HT (high tide) HT+2 (2 hours after high tide) and LT−2 (2 hours beforelow tide)

01-06-96 03-06-96 28-08-96 30-08-96 02-09-97 10-09-97

Depth (m)

LT 5.6 5.7 6.0 4.6 5.7 6.0

LT+2 7.5 7.5 8.0 8.6 6.0 6.5

HT−2 9.5 8.4 9.0 8.6 7.7 7.7

HT 10.0 9.4 9.5 8.8 8.0 8.0

HT+2 – 7.5 7.0 7.0 7.3 7.7

LT−2 8.0 6.1 6.1 6.3 6.3 7.0

Temperature (°C)

LT 17.0 17.0 18.1 17.0 18.7 20.6

LT+2 16.4 16.9 15.6 16.5 18.6 20.3

HT−2 15.3 15.5 15.6 15.5 16.5 17.4

HT 15.5 15.5 15.6 15.4 16.8 17.6

HT+2 – 15.7 15.9 15.5 16.7 17.7

LT−2 16.0 15.9 18.6 16.4 17.3 18.7

Salinity (PSU)

LT 23.7 27.7 35.1 35.4 34.0 33.6

LT+2 30.5 30.1 36.1 35.5 34.2 33.7

HT−2 35.5 35.0 35.7 35.7 35.7 35.5

HT 35.0 35.3 35.8 35.9 35.8 35.6

HT+2 – 35.0 35.7 35.7 35.6 35.5

LT−2 31.0 32.5 35.2 35.4 35.2 35.2

48

Total bacterial abundance varied between 0.2 and8.1 × 109 cells l−1 (Figure 3). Maximal values weremore often found at low tide or 2 hours earlier.

Glucose incorporation Vm (0.48–3.99nmol l−1 h−1) followed a pattern similar to the varia-tion of bacterial abundance with incorporation ratesincreasing towards low tide (Figure 3).

The ranges of variation of the maximum hydroly-sis rate were 189–1662 nmol l−1 h−1 for aminopepti-dase and 1.7–67.0 nmol l−1 h−1 for �-glucosidase

(Figure 3). The tidal pattern of variation was againcharacterised by increasing hydrolytic activity duringebb and a decrease of activity during flood.

Estuarine fluxes

The estimated fluxes of materials and potential bac-terial heterotrophic activities between the estuary andthe ocean during one tidal cycle are represented inFigure 4.

Fluxes of salt were predominantly positive rang-ing from −73 to +727 tons. The net water balancecorresponded to a retention of 2.1–7.7 millions of m3,during a tidal cycle. Negative fluxes of chlorophyll acorresponded to a net export to the sea of 9 to 55 kgduring a tidal cycle. Fluxes of POM were variable insign ranging from −781 to +478 kg per tidal cycle.

Total bacteria and potential ectoenzymatic activi-ties were consistently exported to the coastal sea. Theabsolute values of the net tidal fluxes were 3 to 26 ×1016 cells, 12 to 78 mol h−1 of potential aminopepti-dase activity and 0.3 to 1.1 mol h−1 of potential�-glucosidase activity, per cycle. The net flux of po-tential glucose incorporation was contrasting in signbetween the two 1997 cycles, but higher in spring tide(−0.18 mol h−1 per tidal cycle) as compared to neaptide (+0.02 mol h−1 per tidal cycle).

Discussion

Tidal bacterioplankton dynamics at the estuary-seainterface

Ria de Aveiro is generally described as an estuarinesystem with a complex topography. The location ofstation N1 was defined to allow the best approxima-tion of the physical, chemical and biological proper-ties of the water at the mouth of the estuary since thetotal volume of water that crosses this section corre-sponds to 90% of the total tidal prism (Silva 1994).Since fieldwork at the highly turbulent section be-tween the origin of Canal de Mira and the mouthproved to be impracticable for safety reasons, thecontribution of this channel was not included in thecurrent study.

The patterns of tidal variation of salinity and tem-perature observed at station N1 indicate that daytimewarming occurs as salinity decreases during ebb. Thelagoon has an average depth of 1 m and the shallowsections (<3 m deep) correspond to approximately

Figure 2. Variation of seston, particulate organic matter (POM)and chlorophyll concentrations determined at station N1 duringtidal cycles involving 6 sampling moments: LT (low tide), LT+2 (2hours after low tide), HT−2 (2 hours before high tide), HT (hightide) HT+2 (2 hours after high tide) and LT−2 (2 hours before lowtide). ‰ – 01-06-96; � – 01-06-96; ™ – 28-08-96; © – 30-08-96;� – 02-09-97; � – 10-09-97.

49

95% of the area submersed at high tide (Silva 1994).This allows significant warming of the water columndue to insolation, especially during the warm season.A net export of heat from the estuary to offshore wa-ters was previously reported by Dias et al. (1999).

The concentration of particulate matter did not fol-low a regular pattern over the tidal cycles. This be-haviour independent of salinity has also been reportedfor the variation of DOC in other temperate estuarinesystems and was interpreted as an indication of tid-ally induced resuspension/sedimentation events thatmay exert a major control over the concentration ofparticles in the water column (Miller 1999). However,in the Ria de Aveiro, as in other shallow estuarinesystems, the bed-sediment and the salt marsh tend toretain particles and to export solutes (Newell andKrambeck 1995; Almeida et al. 2002). Inputs of in-organic nutrients and labile DOM may be related topersistence of dense phytoplankton communities andconsequently to the high organic content of sus-pended matter. Assuming that POC corresponds to50% of the POM (Rodier 1996) and using a factor of50 to convert chlorophyll to phytoplankton–C (Epp-

ley et al. 1977), we can estimate that POC corre-sponds to 4–19% of the seston and that phytoplanktonbiomass corresponds to 6–14% of the POC. Thesevalues fit within the ranges usually reported for pro-ductive coastal ecosystems (Charpy et al. 1997; Can-uel and Zimmerman 1999). The higher availability oforganic and inorganic nutrients at the mid- and innersections of the estuary support phyto- and bacteri-oplankton biomasses that are up to 9 times higherthan at the outer section (Cunha et al. 2000). Tidaltransport of cells from upper sections of the estuaryinduced a regular pattern characterised by an ebb-tideenrichment in phyto- (1.3–1.4 fold) and bacteri-oplankton (1.1–10 fold).

Bacterial potential ectoenzymatic activity and glu-cose incorporation also increased at station N1 dur-ing ebb and decreased during flood. However, tidalvariation of activity variables was less pronouncedthan the fluctuation of bacterial abundance. This con-servative tidal fluctuation of activity parameters whencompared with the variation of abundance at stationN1 may result from the anti-parallel variation of totalabundance and specific activity of bacterial cells. Ex-

Figure 3. Variation of total bacterial number (TBN), glucose incorporation Vm, aminopeptidase and �-glucosidase activity determined atstation N1 during tidal cycles involving 6 sampling moments: LT (low tide), LT+2 (2 hours after low tide), HT−2 (2 hours before high tide),HT (high tide) HT+2 (2 hours after high tide) and LT−2 (2 hours before low tide). ‰ – 01-06-96; � – 01-06-96; ™ – 28-08-96; © – 30-08-96; � – 02-09-97; � – 10-09-97.

50

perimental data regarding bacterial communities ofthis estuarine system indicate that bacterioplanktonheterotrophic activity strongly responds to changingwater properties and is negatively influenced by sa-linity (Cunha et al. 2001). According to this scenario,the activity of individual cells and the viability of thecommunity as measured at station N1 decrease dur-

ing ebb as total abundance increases with upper es-tuarine inputs. In contrast, when ocean water is enter-ing the estuary, the density of the community isdecreasing but cells are probably gaining vigourwhile they are being transported to an environmentwith increased nutrient availability.

Figure 4. Net estuarine fluxes of chlorophyll a, particulate organic matter, bacterial cells, potential glucose incorporation, aminopeptidaseactivity, �-glucosidase activity, salt and water through a transverse section of the channel defined at station N1. Net fluxes were estimated asthe difference between the flood tide and ebb tide integrated flows. Shaded bars – Spring tides; black bars – Neap tide

51

Fluxes between the estuary and the coastal oceanicarea

Estuarine impact on the coastal ocean often results inthe formation of plumes of elevated chlorophyll con-centration, bacterial abundance, biomass production,nitrification, and total community respiration (Bian-chi et al. (1999a, 1999b); Malone and Ducklow 1990;Pakulski et al. 2000). The attempt to quantify fluxesof salt and biogenic elements between the Aveiro es-tuarine system and the adjacent Atlantic coast duringthe summer period showed a consistent seaward fluxof bacterioplankton cells and potential heterotrophicactivity.

The magnitude of the exports from the estuary tothe sea reflects the total size and activity of estuarinecommunities and also the balance between marineand freshwater inputs to the main body of the estuaryduring each tidal cycle. The fluxes of bacterial cellsand AMPase activity calculated for the tidal cyclesstudied in 1996 decreased from early to mid summer,when salt imports increased probably as a conse-quence of retention of marine water and reduction infreshwater inputs into the lagoon during the warmseason (Simpson et al. 2001). �-GLCase showed amore irregular temporal pattern. Since this ectoenzy-matic capacity is often reported as being particularlyassociated to low-salinity bacterial communities(Murrell et al. 1999), activation and deactivation pro-cesses of the corresponding enzymes may be super-imposed on the transport effects.

The fluxes of particulate materials at the estuarine-ocean interface are more irregular in sign probablyreflecting physical processes of re-allocation of sedi-ments by tidal and/or residual currents (Cloern andNichols 1985). Maximum current speed at the mouthof the estuary ranges from 0.6 to 2.4 m s−1, (average1.2 m s−1) depending on tidal range and other hydro-graphical conditions (Dias et al. 2000). During thetidal cycle a succession of conditions for resuspen-sion, transport and deposition of particles may occur.Also the frequent disturbance of the channel bed bydredging may interfere with tidal fluxes of suspendedsediments.

The comparison between biogenic estuarine fluxesunder spring and neap tide conditions is not conclu-sive. Results may indicate that other factors, ratherthan tidal range, have a prevailing effect on the mag-nitude of the net estuarine fluxes. Neap tide exports,when compared to spring tide, corresponded to a6-fold increase in chlorophyll, a 32% decrease in bac-

terial cells and a 63% decrease of potential �-glucosi-dase activity. In contrast, neap tide had no significantimpact on the export of potential aminopeptidase ac-tivity, in relation to spring tide values, and coincidedwith a shift in the sign of the fluxes of salt and POM.Short-term variations of the same order of magnitudewere also observed in the 1996 cycles, which corre-sponded to tides of similar range (2.8–3.0 m). Thetwo pairs of cycles separated by 1-day intervals pro-vide a perception of the amplitude of short-term vari-ability not related to changes in tidal range. The re-sults may indicate that other factors controlling thesize and activity of bacterioplankton inside the estu-ary and/or the mechanisms of estuarine water circu-lation exert major influence on the processes of trans-ference between the estuary and the sea.

Conclusion

In spite of the controversy arround Odum’s out-welling hypothesis (Odum 1968) regarding the con-tribution of salt-marsh estuaries to the export of or-ganic matter to the open marine environment (Dameet al. 1986), this study documents transport of bacte-rial communities and increased capacity for the deg-radation of organic matter brought from the shallowestuary of Ria de Aveiro. These exports occurred evenwhen the balances of salt and water indicated a netretention of marine water inside the lagoon enhanc-ing the high levels of activity of estuarine bacteri-oplankton communities, when compared to bacterialassemblages from the open marine environment. Es-tuarine fluxes are highly dependent on the reactivityof bacterioplankton communities and will ultimatelyreflect processes of stimulation and/or deactivation ofheterotrophic capacities during tidal transport alongestuarine gradients.

Acknowledgements

This work was partially supported by PRODEP II,POCTI (Portuguese Foundation for Science and Tech-nology) and CZCM (University of Aveiro, Portugal).Rui Marques, Clarinda Ferreira and a group of under-graduate students helped in field and laboratory workduring tidal cycles.

52

References

Almeida M.A., Cunha M.A. and Alcântara F. 2002. Is bacteri-oplankton production in the Ria de Aveiro influenced by saltmarshes and bed sediments? Aquat. Ecol. 36: 469–482.

Barbosa A. 1991. Spatial and temporal variation of bacterioplank-ton abundance and biomass in a coastal lagoon (Ria Formosa,Southeastern Portugal) Kieler Meeresforsch. Sonderh. 8: 66–73.

Bianchi M., Feliatra and Lefèvre D. 1999a. Regulation of nitrifi-cation in the land-ocean contact area of the Rhône River plume(NW Mediterranean). Aquat. Microb. Ecol. 18: 301–312.

Bianchi M., Fosset C. and Conan P. 1999b. Nitrification rates inNW Mediterranean Sea. Aquat. Microb. Ecol. 17: 267–278.

Billen G. and Garnier J. 1997. The Phison river plume: coastaleutrophication in response to changes in land use and watermanagement in the watershed. Aquat. Microb. Ecol. 13: 3–17.

Canuel E.A. and Zimmerman A.R. 1999. Composition of particu-late organic matter in the southern Chesapeake Bay: sourcesand reactivity. Estuaries 22: 980–994.

Charpy L., Dufour P. and Garcia N. 1997. Particulate organic mat-ter in sixteen Tuamotu atoll lagoons (French Polynesia). Mar.Ecol. Prog. Ser. 151: 55–65.

Cloern J.E. and Nichols F.H. 1985. Time scales and mechanismsof estuarine variability, a synthesis from studies of San Fran-cisco Bay. Hydrobiologia 129: 229–237.

Cloern J.E. 2001. Our evolving conceptual model of the coastaleutrophication problem. Mar. Ecol. Prog. Ser. 210: 223–253.

Cunha M.A., Almeida M.A. and Alcântara F. 2000. Patterns of ec-toenzymatic and heterotrophic bacterial activities along a salin-ity gradient in a shallow tidal estuary. Mar. Ecol. Prog. Ser. 204:1–12.

Cunha M.A., Almeida M.A. and Alcântara F. 2001. Short-time re-sponses of the natural planktonic bacterial community to thechanging water properties in an estuarine environment: ectoen-zymatic activity, glucose incorporation and biomass production.Microb. Ecol. 42: 69–79.

Dame R.F., Chrzanowsky T., Bilstein K., Kjerfve B., Mckellan H.,Nelson D. et al. 1986. The outwelling hypothesis and North In-let, South Carolina. Mar. Ecol. Prog. Ser. 33: 217–229.

Dias J.M., Lopes J.F. and Dekeyser I. 1998. An exploratory studyof the dynamics of Ria de Aveiro, Portugal. In: Kim H., LeeS.H. and Lee S.J. (eds), Hydrodynamics – Theory and Appli-cations. Proceedings of the 3rd International Conference on Hy-drodynamics. Uiam Publishers, Seoul, pp. 619–624.

Dias J.M., Lopes J.F. and Dekeyser I. 1999. Hydrological charac-terisation of Ria de Aveiro, Portugal, in early Summer. Oceanol.Acta. 22: 473–485.

Dias J.M., Lopes J.F. and Dekeyser I. 2000. Tidal propagation inthe Aveiro lagoon, Portugal. Phys. Chem. Earth (B) 25: 369–374.

Dias J.M. 2001. Contribution to the study of the Ria de Aveiro hy-drodynamics. PhD Dissertation, Universidade de Aveiro, Por-tugal.

Eppley R.W., Harrison W.G., Chisholm S.W. and Stewart E. 1977.Particulate organic matter in surface waters off Southern Cali-fornia and its relationships to phytoplankton. J. Mar. Res. 35:671–696.

Fuks D., Devescovi M., Precali R., Krstulovic N. and Solic M.1991. Bacterial abundance and activity in the highly stratifiedestuary of the Krka River. Mar. Chem. 32: 333–346.

Gocke K. 1977. Comparison of methods for determining the turn-over times of dissolved organic compounds. Mar. Biol. 42:131–141.

Hobbie J.E., Daley R. and Jasper S. 1977. Use of Nuclepore filtersfor counting bacteria by fluorescence microscopy. Appl. Envi-ron. Microbiol. 33: 1225–1228.

Hoppe H.G. 1983. Significance of exoenzymatic activities in theecology of brackish water: measurements by means of methyl-lumbeliferyl-substrates. Mar. Ecol. Prog. Ser. 11: 299–308.

Hoppe H.G., Gocke K. and Alcântara F. 1996. Shifts between au-totrophic and heterotrophic processes in a tidal lagoon (Ria deAveiro, Portugal). Arch. Hydrobiol. Spec. Issues Advanc. Lim-nol. 48: 39–52.

Hoppe H.G., Giesenhagen H.C. and Gocke K. 1998. Changingpatterns of bacterial substrate decomposition in a eutrophica-tion gradient. Aquat. Microb. Ecol. 15: 1–13.

Malone T.C. and Ducklow H.W. 1990. Microbial biomass in thecoastal plume of Chesapeake Bay: phytoplankton-bacteri-oplankton relationships. Limnol. Oceanogr. 35: 296–312.

Martin J.M. and Gordeev V. 1984. River inputs to ocean system: areassessment. In: Estuarine processes: an application to the Ta-gus estuary. Proceedings of the Scientific Workshop on Estua-rine Processes. Comissão Nacional do Ambiente, Lisbon, pp.205–240.

Meybeck M. 1998. The IGBP water group: A response to a grow-ing global concern. Global Change Newsletters 36: 8–12.

Miller A.E.J. 1999. Seasonal investigations of dissolved organiccarbon dynamics in the Tamar estuary, UK. Estuar. Coast. ShelfSci. 49: 891–908.

Milliman J.D. and Meade R.H. 1983. World-wide delivery of riversediment to the oceans. J. Geol. 91: 1–21.

Moreira M.H., Queiroga H., Machado M.M. and Cunha M.R. 1993.Environmental gradients in a southern estuarine system: Ria deAveiro, Portugal Implications for soft bottom macrofauna col-onization. Neth. J. Aquat. Ecol. 27: 465–482.

Murrell M.C., Hollibaugh J.T., Solver M.W. and Wong P.S. 1999.Bacterioplankton dynamics in northern San Francisco Bay:Role of particle association and seasonal freshwater flow. Lim-nol. Oceanogr. 44: 295–308.

Newell S.Y. and Krambeck C. 1995. Responses of bacterioplank-ton to tidal inundations of a saltmarsh in a flume and adjacentmussel enclosures. J. Exper. Mar. Biol. Ecol. 190: 79–85.

Odum E.P. 1968. A research challenge: evaluating the productivityof coastal and estuarine water. In: Proceedings of the SecondSea Grant Conference. Univ Rhode Island, pp. 63–64.

Odum E.P. 2000. Tidal marshes as outwelling/pulsing systems. In:Weinstein M.P. and Kreeger D. (eds), Concepts and Controver-sies in Tidal Marsh Ecology. Kluwer Academic Publishers,Dordrecht, pp. 3–7.

Pakulski J.R., Benner R., Whitledge T., Amon R., Eadie B., Cifu-entes L. et al. 2000. Microbial metabolism and nutrient cyclingin the Mississippi and Atchfalaya river plumes. Estuar. Coast.Shelf Sci. 50: 173–184.

Parsons T.R., Maita Y. and Lalli C.M. 1989. A Manual of Chemi-cal and Biological Methods for Seawater Analysis. PergamonPress, Oxford, 172p.

53

Pritchard D.W. 1989. Estuarine classification – A help on ahidrance. In: Neilson B.J., Kuo A. and Brubaker J. (eds), Es-tuarine Circulation. Humana Press, Clifton, pp. 1–38.

Revilla M., Iriarte A. and Orive E. 2000. Bacterial and phytoplank-ton dynamics along a trophic gradient in a shallow temperateestuary. Estuar. Coast. Shelf Sci. 50: 297–313.

Ritzrau W. and Graf G. 1992. Increase of microbial biomass in thebenthic turbidity zone of Kiel Bight after resuspension by astorm event. Limnol. Oceanogr. 37: 1081–1086.

Ritzrau W., Thomsen L., Lara R.J. and Graf G. 1997. Enhancedmicrobial utilization of dissolved amino acids indicates rapidmodification of organic matter in the benthic boundary layer.Mar. Ecol. Prog. Ser. 150: 43–50.

Rodier J. 1996. L’analyse de l’eau: eaux naturelles, eaux résid-uaires, eau de mer. 8th edn. Dunod, Paris, 1383 p.

Silva J.J.F. 1994. Circulação de água na ria de Aveiro: contribuiçãopara o estudo da qualidade da água. PhD Dissertation, Univer-sidade de Aveiro, Portugal.

Simpson J.H., Vennell R. and Souza A.J. 2001. The salt fluxes in atidally-energetic estuary. Estuar. Coast. Shelf Sci. 52: 131–142.

Smith N.P. 2001. Seasonal-scale transport patterns in a multi-inletcoastal lagoon. Estuar. Coast. Shelf Sci. 52: 15–28.

Taylor D.I. and Allanson B.R. 1995. Organic carbon fluxes betweena high marsh and estuary and the inapplicability of the Out-welling Hypothesis. Mar. Ecol. Prog. Ser. 120: 263–270.

Trousselier M., Cahet G., Lebaron P. and Baleux B. 1993. Distri-bution and dynamics of bacterial production in relation to windperturbations in a Mediterranean lagoon. Limnol. Oceanogr. 38:193–201.

von Bodungen B., Gocke K., Smetacek V. and Zeitzschel B. 1995.The effect of sediment flushing by density displacement of in-terstitial water on pelagic primary production and microbialactivity. Universität Kiel, Beitrag Nr. 153 aus dem Sonderfor-schungsbereich 95, p. 87–95.

Yentsch C.S. and Menzel D.W. 1963. A method for the determina-tion of phytoplankton chlorophyll and phaeophytin by fluores-cence. Deep-Sea Res. 10: 221–231.

54