Uranium distribution in the coastal waters and pore watersof Tampa Bay, Florida
Peter W. Swarzenski a,⁎, Mark Baskaran b
a U.S. Geological Survey, 600 4th Street South, Saint Petersburg, FL 33701, United Statesb Wayne State University, Detroit, MI 48202, United States
Received 10 November 2005; received in revised form 3 May 2006; accepted 9 May 2006Available online 27 June 2006
Abstract
The geochemical reactivity of uranium (238U) and dissolved organic carbon (DOC), Fe, Mn, Ba, and V was investigated in the watercolumn, pore waters, and across a river/estuarine mixing zone in Tampa Bay, Florida. This large estuary is impacted both by diverseanthropogenic activity and by extensiveU-rich phosphatic deposits. Thus, the estuarine behavior of uraniummay be examined relative tosuch known U enrichments and anthropogenic perturbations.
processes are initiated in response to fundamental changesin water chemistry as rivers and groundwater mix intoseawater (Sholkovitz, 1976; Boyle et al., 1977; Borole etal., 1982; Carroll and Moore, 1994). Knowledge of theseestuarine biogeochemical transformations is essential tounderstand fully the fate of elements as they are transportedtoward the sea (Millward and Turner, 1994; Swarzenski etal., 1995; Moore et al., 1996). Furthermore, the validity
of estuarine trace-element/radionuclide mass-balance cal-culations is directly bound by our understanding of theseprocesses (McKee et al., 1987; Klinkhammer and Palmer,1991; Shiller and Boyle, 1991; Swarzenski et al., 1995;Andersson et al., 2001). While riverine-flux estimates areusually readily quantifiable, the material contributionderived from submarine groundwater discharge (SGD) isinherently more diffuse and thus much harder to constrain(Porcelli and Swarzenski, 2003; Swarzenski et al., 2003).A study of the uranium geochemistry in Tampa Bayprovides an ideal opportunity to evaluate estuarinebiogeochemical processes in a coastal system influencedby complex surface-water/groundwater interactions andU-rich deposits. Rivers are generally the largest sourceterm for continentally derived weathering products to thesea, and on average contain about 1.3 nM dissolved (tradi-tionally defined as <0.4 μm) and 12.6 nM g−1 particulateU (Moore, 1967; Sackett and Cook, 1969; Mangini et al.,1979; Cochran, 1982; Scott, 1982; Palmer and Edmond,1993). Such global estimates are most useful for large-scale mass-balance derivations and inherently cannotreflect unique seasonal and geographical variations in thedistribution of uranium (Scott, 1982; Palmer andEdmond,1993; Snow and Spalding, 1994). In TampaBay, the combined riverine influx of fresh water is onlyabout 63 m3 s−1 (Weisberg and Zheng, 2006), and thisvalue includes a large contribution (∼ 30%) derived fromnon-diffusive transport through sediments, including hy-porheic exchange and SGD. Such dynamic groundwater/surface-water exchange may impact the delivery and es-tuarine fate of diagenetically sensitive trace elements likeuranium (Lienert et al., 1995).
Riverine trace elements and radionuclides are typi-cally particle reactive and thus largely (∼ 90%) associa-ted with the particulate load of a typical river system(Gibbs, 1977; Martin and Meybeck, 1979; Presley et al.,1980;Martin andWhitfield, 1983; Davis, 1984; Trefry etal., 1986; Zielinski andMeier, 1988; Choppin and Clark,1991; Payne andWaite, 1991; Plater et al., 1992;Waite etal., 1994; Lienert et al., 1995; Swarzenski et al., 1995).These ionized riverine particles/colloids can be efficient-ly stripped of their trace elements/radionuclides duringestuarine mixing in response to biogeochemical reac-tions initiated by an increase in ionic strength (Sholk-ovitz, 1976, 1977; Boyle et al., 1977;McKee et al., 1987;Swarzenski et al., 1995; Moore et al., 1996).
In addition to such fluvial-source terms, much recentevidence (Burnett et al., 2003) indicates that SGD mayalso contribute substantively to estuarine mass budgets,particularly along coastlines that do not have largedischarging rivers. It is important to recognize thatSGD need not constitute fresh groundwater, but rather
a composite of recycled seawater, as well as meteoricand connate groundwater. Reactions and processes duringSGD may be comparable to those that occur during hy-porheic exchange. For example, diagenetic transforma-tions within the seabed and associated pore waters (orgroundwater) may impact the estuarine U behavior byserving as either a sink or potential source forU, dependingon the redox state and carrier phase (Cochran et al., 1986;McKee et al., 1987; Anderson et al., 1989; Barnes andCochran, 1990, 1993; Shaw et al., 1995; Swarzenski et al.,1995, 2004). Dissolved U in pore waters can be mobilizedeither by Fickian diffusion across the sediõment/waterinterface (Barnes andCochran, 1993) or by advective fluidtransport mechanisms, such as SGD or hyporheic ex-change. The release of U from particles or colloids thathave already undergone partial diagenetic alterations isthought to be an additional source for reactive U into thewater column (McKee et al., 1987; Swarzenski et al.,1995). Another important source of U to Tampa Bay maybe derived from the ubiquitous phosphogypsum deposits.Although the uranium-rich Miocene phosphatic depositscontained within the Bone Valley Member of the Haw-thorn Group (Osmond et al., 1984) of west-central Floridaare not unique globally, environments where such depositsare actively forming today are unknown (Green et al.,1995). The unique U concentration and isotopic compo-sition of the phosphatic deposits may enable U and itsnatural isotopes (234,238U) to possibly be utilized as uniquegroundwater mass tracers into estuarine waters under idealconditions (Osmond and Cowart, 1976).
Lastly, it is well known that U can be highlyenriched and can show extreme isotopic disequilibriumin groundwater (Osmond and Cowart, 1976; Fleischerand Raabe, 1978; Hussain and Krishnaswami, 1980;Copenhaver et al., 1993; Snow and Spalding, 1994;Porcelli and Swarzenski, 2003). In areas where coastalgroundwater is actively discharged into seawater(Moore, 1996), it may also be possible to separatethe isotopic U activity within such inflowing ground-water from oceanic and fluvial isotopic signatures. Acomprehensive mass balance of U and its daughters incoastal waters should consequently include an evalu-ation of such coastal groundwater discharges. Unfor-tunately, quantitative assessment of submarinegroundwater discharge and associated trace-elementand radionuclide fluxes into coastal waters is stilldifficult to resolve on a regional scale.
This report addresses the seasonal variability of dis-solved uranium and dissolved organic carbon (DOC),Fe, Mn, Ba, and V in the surface and pore waters ofTampa Bay. Our results indicate that the role of fluidtransport across the sediment/water interface, including
45P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
SGD and hyporheic exchange, has a controlling effect onthe estuarine geochemistry of uranium.
2. Physiographic setting
Tampa Bay is situated along the west-central coast ofFlorida and has historically been classified as a partiallyto well-mixed drowned river-valley estuary. A recent re-evaluation of the geologic history of Tampa Bay suggeststhat this bay instead likely developed from a series ofkarst-related collapse features in response to sea-levelfluctuations (Brooks and Doyle, 1998). The Tampa Baywatershed and bay surface areas are about 5700 km2 and1030 km2, respectively, whereas the mean depth is quiteshallow (∼ 4 m), yet can extend to ∼ 15 m within navi-gational channels (Weisberg and Zheng, 2006). The totalvolume of the bay is about 4×109 m3, and a yearlyaverage riverine inflow of 63 m3 s−1 is divided amongthe Hillsborough River (23.8%), the Alafia River(20.6%), the Little Manatee River (9.5%), and the Ma-natee River (15.9%). The remaining∼ 30% of the annualriverine inflow consists of ungauged influxes to the bay,such as those from small streams, hyporheic exchange,and direct runoff.
3. Field and laboratory methods
3.1. Surface waters
A salinity transect across the Alafia River/estuarinemixing zone was sampled on June 15, 2001 (Table 1).Subsequently, a series of 17 stations throughout TampaBay was sampled (Fig. 1) repeatedly for uranium (238U),dissolved organic carbon (DOC), Fe, Mn, Ba, and Vanda suite of ancillary constituents during June 2003, Au-gust 2003, and June 2004; each cruise corresponded tounique hydrologic conditions as shown in the bay-widesalinity values listed in Tables 2 and 3 and Fig. 3.
Table 1Geographic position, salinity, and concentrations of U, V, and Ba across an
Station Latitude Longitude S
AR0 27° 51.601 N 82° 16.475 W 0AR2 27° 52.179 N 82° 17.108 W 2AR4 27° 52.512 N 82° 17.496 W 4AR6 27° 52.872 N 82° 18.688 W 5AR8 27° 52.200 N 82° 18.795 W 7AR9 27° 51.979 N 82° 18.761 W 8AR10 27° 52.119 N 82° 19.362 W 13AR14 27° 51.876 N 82° 20.049 W 18AR25 27° 51.446 N 82° 21.369 W 24AR31 27° 51.545 N 82° 23.072 W 30AR33 27° 50.855 N 82° 25.841 W 32
Water samples were collected at each station usingmethods to ensure trace-metal clean procedures. About1 L of water was drawn from a depth of 0.5 m using aperistaltic pump and Teflon tubing attached to a gim-baled sampling port. The water was immediatelyfiltered in situ through pre-cleaned 0.4-m GeoTechfilter cartridges. All filtration equipment and collecting/receiving vessels were acid cleaned and triple rinsedwith Milli-Q water prior to use. The dissolved trace-element fractions were immediately acidified to pH 1–2 with double-distilled SeaStar HCl and stored in adark/cool environment for subsequent processing;nutrient splits were stored on ice on board and frozenin the lab prior to processing. A separate volume ofwater (∼ 1 L) was collected using a dark bottle andstored chilled for subsequent gravimetric suspendedparticulate matter (SPM) analyses. Surface-watercolumn DOC samples were collected using a metalbucket and immediately filtered on board through pre-combusted GFF filters using a clean glass-filtrationapparatus.
3.2. Pore waters
Interstitial waters were collected from various depths(<2 m) at select sites (n=4) within Tampa Bay usingeither a multi-sampler (Martin et al., 2003) or a singlescreened drive point piezometer. Dissolved oxygen aswell as other standard hydrographic parameters weremonitored continuously prior to sample collection toensure contamination-free samples. After sufficient equil-ibration (dissolved oxygen concentrations approachedzero), small volumes of pore water (<250 ml) were drawnfrom individual horizons using a sampling systemconsisting of a small volume peristaltic pump, acid-cleaned tubing and an inline 0.4-m filtration cartridge.Such a system enables the careful collection of redoxsensitive chemical species in pore waters.
Alafia River salinity gradient (June 2001)
alinity U (nM) V (nM) Ba (nM)
.0 3.6 131 64.4
.2 2.2 55 16.6
.0 3.2 82 68.7
.9 3.7 71 77.5
.8 5.0 74 84.5
.9 5.6 73 82.8
.5 9.1 95 99.7
.6 13.6 89 112.3
.7 14.8 81 129.1
.6 16.7 59 130.4
.8 16.2 51 122.1
Fig. 1. Station location map for Tampa Bay showing both pore water and water column sites; all surface water samples were collected during June2003, August 2003, and June 2004.
The salinity, temperature, pH, dissolved oxygen, andspecific conductance for each water column and porewater sample were obtained using a calibrated YSI CTD
probe or comparable device. Trace-element concentra-tions were obtained using a sector field ICP-MS (Fin-nigan MAT Element 2) at the University of SouthernMississippi, as per methods of Rodushkin and Ruth(1997). Samples were first diluted 10-fold with 0.1 M
Table 2Geographic position, depth, salinity, and concentrations of suspended particulate matter (SPM), U, V, Ba, and DOC at select sites within Tampa Bayduring June 2003
Station Latitude Longitude Depth (m) Jun-03
Salinity SPM (mg L−1) U (nM) V (nM) Ba (nM) DOC (mM)
TB01 27° 53.7825 N 82° 27.8517 W 2.87 23.10 9.47 9.8 59.6 35.0 369TB02 27° 52.9617 N 82° 25.2828 W 3.11 21.60 10.00 9.4 66.7 48.7 418TB03 27° 39.9881 N 82° 31.8785 W 4.69 28.40 8.78 10.4 44.8 33.8 283TB04 27° 45.6011 N 82° 34.5732 W 3.87 26.00 5.22 10.5 58.0 46.7 311TB05 27° 41.6805 N 82° 32.6138 W 3.08 28.00 2.86 10.7 46.8 39.8 291TB06 27° 43.3730 N 82° 30.8804 W 3.60 27.70 3.29 11.2 49.4 25.1 277TB07 27° 59.0500 N 82° 39.3959 W 2.90 22.10 8.60 9.6 72.9 79.0 387TB08 27° 56.5204 N 82° 40.1675 W 4.42 21.50 13.57 9.7 75.3 76.0 407TB09 27° 57.0734 N 82° 35.9517 W 5.39 22.50 6.17 9.9 67.0 56.5 384TB10 27° 53.4133 N 82° 35.8801 W 2.59 24.10 5.87 10.4 63.3 91.4 –TB11 27° 52.7385 N 82° 33.7175 W 5.21 25.20 6.27 10.5 60.9 66.1 –TB12 27° 47.0576 N 82° 26.6938 W 4.48 24.50 10.93 10.4 58.6 74.1 427TB13 27° 46.6942 N 82° 30.3180 W 7.77 25.20 6.27 10.3 57.2 76.0 409TB14 27° 48.2771 N 82° 33.3144 W 4.02 25.20 8.13 10.1 59.1 62.6 480TB15 27° 43.3821 N 82° 35.6567 W 3.90 27.40 6.27 10.5 48.3 26.6 381TB16 27° 37.0038 N 82° 31.9600 W 3.38 29.40 2.53 11.0 45.3 27.5 459TB17 27° 51.5872 N 82° 27.2072 W 4.85 23.00 10.40 10.0 64.8 58.9 455
47P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
HNO3 and analyzed versus aqueous standards similarlydiluted with 0.1 M HNO3. Each of the six elements wasdetermined in triplicate. Briefly, a small (ml) aliquot ofeach sample was spiked with 235U and diluted accordingto salinity with 1% HNO3. In the case of uranium, the235U/238U ratio was determined with a timed-intervalscan of the ICP-MS. Detailed analytical methods can befound in Swarzenski et al., (1995, 2001). Such a methodfor U detection was highly reproducible and accurate:10 replicate analyses of a standard (968 ng L−1) pro-duced a mean±2 ng L−1 from a spiked standard with a
Table 3Salinity and concentrations of SPM, U, V, Ba, and DOC at select sites withi
standard deviation of 15 ng L−1. At these concentra-tions, the precision of the data was typically±1.5%.
4. Results and discussion
4.1. Alafia River transect
The Alafia River is one of the largest free-flowingrivers that discharges into Tampa Bay and originates, atleast partially, about 30 km upstream at Lithia Springs.Discharge in the Alafia River (Dooris and Dooris, 1985)
fluctuates from a dry-season low of about 5.4 m3 s-1 to28.3m3 s-1 (average=14m3 s-1). Groundwater of varyingage and compositionmay flow either directly into the riverat known spring sites or may seep into river bottom wateras baseflow under favorable hydraulic gradients. Distin-guishing river water from groundwater in such coastalrivers is thus often complicated by bi-directional ground-water/surface-water exchange. Classic estuarine transfor-mation reactions, triggered among other processes by anincrease in salinity, will thus likely also be influenced bythe redox state of this readily exchanged water.
Select trace-element (V, Ba, and U) concentrationsfrom the Alafia River salinity transect are tabulated inTable 1 and presented in Fig. 2. Relative to a conserva-tive mixing line, V exhibits pronounced removal fromthe water column across a salinity range from 1 to 9.Removal of V is likely tied to the reactivity of its domi-nant carrier phase, either Mn- or Fe-oxyhydroxides ororganic surfaces (Shiller and Boyle, 1987, 1991). Incontrast, Ba behaves somewhat non- traditionally in thisriver/estuary mixing system by exhibiting conservativebehavior from a salinity of 1 to about 9, and only athigher salinities shows classic water column enrichmentdue to particle-desorption phenomena (Li and Chan,1979). Uranium may be introduced to the Alafia Riversystem in several ways: i) during erosion of U-bearingstrata such as limestone and clays; ii) seasonal runoff oforganic, reducing peatbogs; iii) groundwater infiltration;and possibly iv) as a by-product of large-scale fertilizer
Fig. 2. Concentrations of (A) U, (B) V, and (C) Ba across a salinitygradient in the Alafia River/Tampa Bay mixing zone.
(i.e., PO4−3) mining activities (Osmond and Cowart,
1976; Mangini et al., 1979). Uranium shows both remo-val and enrichment across the salinity gradient; removalfrom the water column is evidenced by a salinity of 1 to9, and enrichment is observed beyond a salinity of 15.Such a non-conservative U:salinity relation has beenpreviously observed in other estuaries, including thecolloid-rich Amazon River mixing zone (Swarzenski etal., 1995, 2004) and across the O2/H2S boundary in apermanently anoxic fjord (Swarzenski et al., 1999).Considering that the Alafia River is not significantlyparticle-laden and is complexly involved in reversiblegroundwater/surface-water exchange processes (Plateret al., 1992; Lienert et al., 1995), the reactivity of U islikely influenced or controlled by the fractional contri-bution derived from groundwater, the redox state of thisfreshly introduced water, as well as the nature of organicbinding sites (Davis, 1984; Zielinski and Meier, 1988;Mann and Wong, 1993). Uranium concentrations in ex-cess of average seawater values (13.6 nM; Chen et al.,1986) observed at Stations AR25, AR31, and AR33 in-dicate an additional, non-riverine source.
4.2. Water column
The water column composition of Tampa Bay is in-fluenced by precipitation (i.e., storms), evaporation, un-gauged and gauged river inflows, wetland-exchangeprocesses, and the bidirectional exchange of coastalgroundwater across the sediment/water interface. Due tothe relatively small watershed size, each of these drivingcomponents is directly a function of the local seasonal orlonger-term climatic variability. Submarine groundwaterdischarge in this systemmay include hyporheic exchangewithin coastal streams/rivers, surficial aquifer-exchangeprocesses around the periphery of the bay, as well asgroundwater discharge from the Floridan Aquifer system.Water column salinity values can fluctuate widely inTampa Bay and often reflect very recent precipitationevents rather than predictable seasonal signatures.
Concentrations of suspended particulate matter (SPM),salinity, U, V, Ba, and dissolved organic carbon (DOC) insamples collected during June 2003, August 2003, andJune 2004 are presented in Tables 2 and 3. One of thestriking features of these datasets is that a full salinityregime as evidenced in most other estuaries is not usuallyobservable in the water column of Tampa Bay. Rather, thesalinity fields are typically skewed toward more marinevalues (Fig. 3, top). To take advantage of an unexpectedand highly uncharacteristic freshened water column,which occurred as a result of unusually heavy, localprecipitation rather than an anomalous river-discharge rate,
Fig. 3. Bay-wide contour plots of water column salinity (top) and U concentrations (bottom) measured at select sites in Tampa Bay during June 2003, August 2003, and June 2004.
49P.W
.Sw
arzenski,M.Baskaran
/Marine
Chem
istry104
(2007)43–57
Fig. 4. (A) Salinity versusU concentration and (B) suspended particulatematter (SPM) plots from select siteswithin TampaBay during June 2003,August 2003, and June 2004. Note the average seawater (13.6 nM; Chenet al., 1986), the average global riverine U concentration (1.3 nM; Palmerand Edmond, 1993), and an average local riverine U concentration(6.1 nM; Table 4) are shown for comparison.
the water column of Tampa Bay was sampled in August2003, just two months after the June 2003 sampling.Average salinity values observed at stations 1–17 duringJune 2003, August 2003, and June 2004 were 25.0, 16.3,and 28.1, respectively, while the lowest salinity (10.7)observed bay-wide during all three sampling effortsoccurred at station 1 in August 2003.
Several factors contribute to such elevated salinityfields in Tampa Bay: i) hydraulic gradients in the small,spring-fed watersheds adjacent to Tampa Bay are insu-fficient to produce significant river discharge; ii) the watercolumn residence time in the bay is thought to be rela-tively long (Tr>weeks; Swarzenski et al., 2007-thisvolume), and much of the bay is very shallow, enablingenhanced evaporation; and iii) the composition ofgroundwater beneath Tampa Bay is mostly salty, suchthat the discharge of most submarine groundwater cannotprovide a mechanism for water column freshening.
While these coastal rivers do not contribute largefluxes of water to the bay, they also do not transportmuch suspended particulate matter. Suspended particu-late matter (SPM) concentrations (operationally definedhere as the fraction collected on a 0.4-m Nucleoporemembrane) in the bay ranged (Fig. 4B) from just 2 to14 mg L−1 (June 2003) and from 1 to 10 mg L−1 (June2004). Due to the paucity of inflowing riverine particles,it is likely that these low SPM concentrations reflectinstead either natural or ship-related resuspension eventsand/or localized algal blooms.
Considering that the Alafia River freshwater end-member U concentration (3.6 nM) is∼ 3 times the globalriverine U value of 1.3 nM (Bhat and Krishnaswami,1969; Mangini et al., 1979; Palmer and Edmond, 1993),the observed estuarine U concentrations in TampaBay arecomparably 2 to 3 times greater than those found in mostother estuarine systems (McKee et al., 1987; Sarin andChurch, 1994; Somayajulu, 1994; Carroll and Moore,1994; Church et al., 1996; Swarzenski et al., 1995, 2003,2004; Swarzenski and McKee, 1999). Concentrations ofU as high as 16.7 nM (AR31, Table 1) revealed in the highsalinity reach of the Alafia River estuary suggest ad-ditional source(s) of U within this region (Osmond andCowart, 1976). A likely source that could yield such highwater column concentrations is related to either the phos-phate-mining operations at the mouth of the Alafia Riveror to the flux of dissolved U from pore waters across thesediment/water interface. For example, within an ap-propriate hydrogeologic framework, it is likely that suchflux rates could be greatly augmented during widespreadSGD or bio-irrigation. Some geochemical aspects ofwidespread bio-irrigation in Tampa Bay are examined inKlerks et al. (2007-this volume) who reported that
abundant, deep-burrowing ghost shrimp contribute tothe spatial heterogeneity of select sedimentary traceelement concentrations, and that ghost shrimp bioturba-tionmay result in a significant flux of trace elements to thesediment surface. The potential impact of SGD on theestuarine U geochemistry in Tampa Baywill be addressedin a subsequent section of this paper.
Fig. 4A shows the distribution of U as a function ofsalinity during all three bay-wide sampling efforts. Thewell-defined open ocean U concentration of 13.6 nM, asestablished by Ku et al. (1977) and Chen et al. (1986), theaverage global riverine U value (1.3 nM; Palmer andEdmond, 1993) and an average local (6.1 nM=(5.3+ 6.9) /2; Table 4) riverine U concentration are also plotted forcomparison. It is evident that during June 2003 and June2004when salinitieswere consistently in excess of 20 bay-wide, observed estuarine U variability was much less thanduring August 2003. During the recently freshened watercolumn (salinity>10.7) observed in August 2003, Uconcentrations were much more variable as a function ofsalinity. Since the greatest U concentrationswere observedat salinities close to 15 and not 10, it is not likely that suchnon-conservative U behavior can be attributed to pulseriver-discharge events. Instead, we suggest that U-richpore water is more efficiently exchanged across thesediment/water interface at select sites during periods ofhigher precipitation (Shaw et al., 1995).
Dissolved organic carbon (DOC), Ba, and Vare plot-ted as a function of salinity during the three sampling
Fig. 5. (A) Water column dissolved organic carbon (DOC), (B) Ba, and(C) V concentration versus salinity plots measured at select sites withinTampa Bay during June 2003, August 2003, and June 2004.
Table 4Concentrations of 238U (nM) and 234U/238U activity ratios in surface- and groundwater samples within Tampa Bay and adjacent rivers
Site Date Latitude Longitude Salinity U (nM) 234U/238U
Alafia River 4/15/2004 27° 52.1100 N 82° 19.3620 W 0.24 5.34±0.17 1.144±0.009Little Manatee River 4/13/2004 27° 40.2480 N 82° 21.1680 W 0.18 0.69±0.01 1.330±0.050Hillsborough River 4/14/2004 28° 1.1820 N 82°27.0240 W 1.66 6.89±0.17 0.930±0.002Tampa Bay I 1/19/2005 27° 51.5872 N 82° 27.2072 W 27.3 5.80±0.21 1.126±0.001Tampa Bay II 4/13/2004 27° 36.4980 N 82° 34.2840 W 29.04 10.6±0.4 1.140±0.001Feather Sound well 4/14/2004 27° 54.8040 N 82° 39.6000 W 11.54 66.4±1.3 1.086±0.003Bullfrog Creek well 4/15/2004 27° 50.2728 N 82° 23.8038 W 21.31 1.09±0.04 1.098±0.011Phosphate plant effluent 4/10/2004 27° 37.4172 N 82° 31.8877 W 0.1 252.5±0.3 1.011±0.002
51P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
efforts in Fig. 5A, B, C, respectively. A June 2003 andAugust 2003 DOC:salinity distribution suggests againthat the variability in DOC, which shows non-conserva-tive behavior, is introduced during the wetter (August2003) sampling effort. Such a distribution implies thatthe DOC, which may also exist in a colloidal phase, isstill relatively labile or utilizable during estuarine mixingprocesses and may be newly introduced to the bay, eithervia riverine transport or SGD. From a salinity of about 15to 30, DOC concentrations approached ideal mixing andindicate eventual transformation into more refractoryforms of carbon (Sholkovitz et al., 1978). Although Umay have an affinity for organic surfaces in natural wa-ters (Li et al., 1980; Zielinski andMeier, 1988; Payne andWaite, 1991; Mann and Wong, 1993), in Tampa Bay aplot of U versus DOC does not indicate a strong cor-relation (r2 =0.2).
Barium is a trace element that is actively involved inbiological cycles within marine systems (McManus et al.,1998). Non-conservative behavior of dissolved Ba inriver-dominated estuaries is well documented (Coffey etal., 1997) and has been attributed to seawater induced ionexchange reactions (Hanor and Chan, 1977). The extentof non-conservative behavior is influenced by physicalmixing kinetics. Fig. 5B suggests that the estuarine Badistribution is highly dependent on the individual sam-pling episode; each sampling effort produced unique Ba:salinity relations. For example, the linear Ba distributionas a function of salinity in Tampa Bay observed duringJune 2004 appears to indicate conservative mixing. Incontrast, Ba exhibited much more variability during June2003 and June 2004. It is likely that the reactive nature ofBa reflects unique interactions with localized biologicalcycles (i.e., algal blooms or microorganisms). Barium hasalso been shown to be a potentially useful tracer of SGD(Shaw et al., 1998) andmay be enhanced in the pore fluidsof coastal aquifers. Of the three studied trace elements, Vmixedmost conservatively with salinity during June 2003and June 2004. The variability in V concentrations wasgreatest during August 2003 and in those samples with
salinities just above 10. Such behavior again may beattributed to source variability during local rain events,when the discharge from rivers and SGD are also en-hanced. The geochemistry of V has been previouslylinked to water column redox cycling in the MississippiBight region (Shiller and Mao, 1999). Unlike the othertrace elements, V may also have an atmospheric-sourceterm, and that signature would be expressed most duringwet periods (Shiller and Boyle, 1991).
4.3. 234U/238U activity ratios (UARs)
Concentrations of 238U and activity ratios of 234U/238Ufrom a suite of surface- and groundwater samples arepresented in Table 4. If groundwater that is exchangedwith estuarine water has a unique 234U/238U activity ratio
compared to ocean- (1.14; Chen et al., 1986) and river-water, one could potentially use 234U/238U activity ratiosto quantify rates and locations of submarine groundwaterdischarge. Uranium activity ratios have been used simi-larly to distinguish various groundwater masses and toconstruct groundwater age models (Osmond and Cowart,1976). Surface waters from the three inflowing riversindicated considerable variability in both the U concen-trations and UARs. Such fluctuations likely representunique geologic controls that define the spatial distribu-tion of uranium in Tampa Bay. While the concentrationsofU in the twowellwater samples also showed significantgeographic variability, the UARs from these two sampleswere virtually indistinguishable.As expected, UARs fromthe two bay sites approached the oceanic endmembervalue, indicating a strong marine control on the isotopicbehavior of uranium in Tampa Bay. One water samplecollected downstream from a phosphate processing plantwas highly elevated inU (252 nM), but was isotopically inequilibrium. Although our preliminary UAR resultssuggest that the coastal groundwater observed in TampaBay is isotopically not unique to both bay and river water,we cannot entirely discount the utility ofUARs as an SGDtracer. Extensive recycling of seawater within a coastalaquifer would yield similar UAR results. Clearly, moredata are needed to establish the utility of uranium anduranium isotopes as congruent tracers of submarinegroundwater discharge.
4.4. Pore waters
Chemical constituents present within pore waters aresensitive indicators of post-depositional diagenetic change(Aller and Cochran, 1976; Froehlich et al., 1979; Cochranet al., 1986; Aller et al., 1986; Nagao et al., 1992). Porewater dissolved iron— Fe(II), manganese—Mn(II), Ba,V, and U concentrations from four sites, in addition torespective overlying bottom waters, are presented inTable 5 and illustrated in Fig. 6. Each pore water site wasselected based on expected variations in the underlyinghydrogeologic setting and proximity to the two phosphate-industry distribution ports. As a result of the widespreadreductive dissolution of Fe oxides present in sedimentsunderlying Tampa Bay, a sharp increase in dissolved Fe(II) was consistently observed at a depth of about 10 cm ateach porewater station. This zone of elevated dissolved Fe(II) ranged from 7535 nM at station MD1 to 22673.8 nMat BH2 and was broadest (688 nM at 50 cm) at AR1. Abimodal distribution in Fe(II) was observed at BH2, wherea secondary Fe peak (9785 nM) occurred at a depth of80 cm. Similar to the observed down core pore water Fe(II) distributions, the most pronounced spike in dissolved
Mn(II) also occurred at depths of about 10 cm. However,whereas Fe(II) concentrations typically decreased rapidlybelow this depth, a more gradual decrease in pore waterMn(II) concentrations at select sites (e.g., PS1) indicatesan expanded Mn oxidation/reduction zone. While a likelyconsequence of such extensive iron/manganese reductionin the absence of significant sulfide production wouldresult in the formation of authigenic minerals (e.g., ura-ninite, calcite, siderite, vivianite, and rhodocrosite), ob-served elevated pore water sulfide concentrations(Swarzenski, unpublished data) render these mineralsunstable (Aller et al., 1986, 1996). The similarity betweenthe Fe and Mn pore water profiles indicates that the dia-genetic sequence is influenced foremost by slow sedi-mentation rates and the rapid reduction of SO4
−2 and theassociated formation of sulfides.
The redox chemistry of uranium in pore waters ofmarine sediments has been studied extensively to betterunderstand its removal mechanism from seawater intosediments (Anderson et al., 1989; Klinkhammer andPalmer, 1991; Nozaki, 1991; Nagao et al., 1992; Barnesand Cochran, 1993; Sarin and Church, 1994; Shaw et al.,1995; Church et al., 1996; Swarzenski et al., 2004). Inneutral to slightly alkaline oxygenated waters, uraniumexists as a stable, hexavalent, uranyl carbonate complex,UO2(CO3)3
4−, while under reducing conditions, U(VI) isreduced to a much less soluble U(IV) form (Langmuir,1978). In seawater, this redox transformation occurs closeto the electron potential necessary for the reduction of Fe(III) to Fe (II) (Langmuir, 1978; Hsi and Langmuir, 1985;Waite et al., 1994). Thermodynamic models predict thatporewater U in anoxic sediments dissolving Fe(II) and S−
in pore water should be reduced and thus removed fromsolution (Barnes and Cochran, 1990, 1993; Swarzenski etal., 2004). Previous studies, however, show that the porewater U concentrations often exceed the overlying sea-water U value (Kolodny and Kaplan, 1973; Boulad andMichard, 1976; Zhorov et al., 1982; Cochran et al., 1986).
The pore water uranium distribution profiles in TampaBay exhibit considerable inter-site variability as evidencedby subsurface U maxima (at sites AR1, MD1), asubsurface minimum (PS1), and also broad zones ofoxidation/reduction (BH2). Consequently, depending onthe observed concentration gradient, U may at times(AR1, MD1) diffuse out of sediments into bottom waters,or conversely, flux into sediments (PS1, BH2) if the watercolumn U concentration is greater than the pore water Uconcentration just below the sediment/water interface. AtAR1, U concentrations increased from a bottom watervalue of about 17.6 nM to 24 nM at depths of 10 and30 cm. Such U concentration gradients imply a net flux ofU upward across the sediment/water interface. We
Table 5Pore water salinity, dissolved oxygen (DO; mg L−1), Fe, Mn, Ba, V, and U concentrations (nM) at select sites in Tampa Bay
Site Depth (cm) Salinity DO (mg L−1) Fe (nM) Mn (nM) Ba (nM) V (nM) U (nM) Jdiff (mol d−1) JSGD (mol d−1)
A diffusive uranium pore water flux rate (Jdiff, mol d−1) is calculated assuming a molecular-diffusion coefficient of 2.7×10−6 cm2 s−1 and a porosityof 80%.Negative fluxes indicate upward exchange across the sediment/water interface. Also shown is an estimated submarine groundwater discharge derivedU flux (JSGD, mol d−1) calculated from the bay-wide Ra-derived SGD rate of 8.3 L m−2 d−1 (Swarzenski et al., 2007-this volume).
53P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
calculated a diffusive flux (Jdiff) of pore water uraniumfrom the four sites (Table 5) using a derivation of Fick'sfirst law, and assuming a linear concentration gradientbetween the uppermost pore water data point (usually 10-cm depth) and a bottom water value, so that
Jdiff ¼ −/Ds∂C∂z
� �z¼0
nM cm−2 y−1� �
; ð1Þ
where Ds is the molecular diffusion coefficient (cm2 s−1)for U, corrected for tortuosity, is the average sedimentporosity (cm3 water per cm3 sediment) of the uppermostinterval, and C is the uranium concentration at a depth z.The value of Ds, 2.7×10
−6 cm2 s−1, is derived using avalue tabulated for the uranyl ion (Li and Gregory, 1974)corrected for tortuosity (Berner, 1980) using a porosityvalue of 80%. In this approach, we have assigned a ‘best-fit’ gradient (δC/δz) based on the sampling resolution, andestimate an uncertainty in the calculation of Jdiff of ±25%,due to uncertainties in the diffusion coefficient calculationand in the depth assignment. By assuming a linear con-
centration gradient, the resulting flux would be a con-servative or lower estimate.
Values of Jdiff ranged from −82 to −57.7 mol d−1 atsites AR1 and MD1, indicating an upward flux ofuranium out of the sediments, to 15.5–116.6 mol d−1 atPS1 and BH2, respectively.
Such positive fluxes imply removal of water columnUconcentrations into the sediments at sites BH1 and PS1.The calculated upward flux of pore water U at AR1 andMD1 must be of sufficient magnitude to drive the bottomwater U concentrations to exceed average seawater Uconcentrations (13.6 nM). For comparison, a bay-wide,Ra-derived submarine groundwater discharge estimate forTampa Bay (8 L m−2 d−1; Swarzenski et al., 2007-thisvolume) yielded advective (JSGD) U fluxes that rangedfrom 82.9 to 203.2 mol d−1, with an average of 112.9 mold−1. The above calculated U fluxes suggest that the porewaters or coastal aquifer underlying Tampa Bay mostlyprovide a net source of U to the bay, although there is alsoevidence for site specific U removal. Such findings areconsistent with the karstic nature of this estuary.
Fig. 6. (A) Pore water Mn(II), Fe(II), 238U, and (B) V, and Ba concentrations at sites AR1, MD1, PS1, and BH2 within Tampa Bay.
Pore water vanadium profiles were often similar inshape to U pore water distributions. At all sites, baybottom water V concentrations were consistently in ex-cess of pore water concentrations, which suggests thatvanadium is removed from the water column to thesediments (Shiller and Mao, 1999). There is no apparentevidence for coincidence between pore water V and Fe/Mn cycling. Pore water barium concentrations at all sitesexceeded bottomwater values, indicating a consistent fluxof Ba from the sediments into bay bottom waters. If pore
water Ba concentrations were controlled only byequilibrium reactions with marine barite, then onewould expect that the Ba pore water concentrationswould increase monotonically with depth, approachingsome asymptotic concentration. This, however, is notobserved in the pore waters of Tampa Bay. Instead, at allsites, the observed Ba pore water maximum is coincidentwith the Fe or Mn maximum, possibly reflecting aninteraction between Ba and metal-oxide cycling. Forexample, under such a scenarioBa could be adsorbed onto
55P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
Mn oxides during the oxidation of reduced Mn(II) andsubsequently released during reduction. At depths greaterthan 10 cm, Ba profiles followed a down-core decrease inconcentration. One possible interpretation of such Babehavior is that a phase with high solubility dissolves inthe upper sediments, while precipitation of a less solublephase occurs at depth. Such an interpretation would implyat least two distinct depositional sequences present bay-wide. Alternatively, energetic pore water bio-irrigation(Klerks et al., 2007-this volume) or some other non-diffusive transport process such as SGD could alsocontribute to these Ba distributions (Shaw et al., 1998).
5. Summary
1) Uranium exhibited both removal and enrichmentacross the Alafia River estuarine mixing zone. Suchnon-conservative U behavior may be attributed to: i)the carrier phase of uranium, which in the absence ofsignificant terrigenous suspended material may be or-ganic in nature; and ii) dynamic groundwater/surface-water (hyporheic) exchange processes ubiquitous inFlorida coastal rivers that may control U removal andenrichment through U(IV)–U(VI) valence statetransformations.
2) Bay-wide uranium concentrations were∼ 2 to 3 timesgreater than those reported for other estuarine systemsand are likely a result of erosional processes of theubiquitously present U-rich phosphatic deposits of theHawthorn Formation. Although estuarine dissolvedU-concentrations generally did not approach seawatervalues (13.6 nM; Chen et al., 1986) along the salinitytransect, in select regions of the bay, water column Uconcentrations did exceed 16 nM. Within the hydro-geological framework of the bay, such enriched Umaybe derived from a enhanced transport processes acrossthe sediment/water interface, such as SGD or bio-irrigation.
3) Pore water profiles of U in Tampa Bay show both aflux into and out of bottom sediments, depending onthe site location and underlying hydrogeologic frame-work. First order diffusive (Jdiff) uranium pore waterflux rates ranged from −82 mol d−1 (AR1) to 116.6(BH2) mol d−1, while Ra-derived SGD estimatesproduced advective U fluxes (JSGD) that ranged from82.9 mol d−1 (BH2) to 203.2 mol d−1 (AR1). Thereappears to be a consistent coincidence of dissolved Uwith Fe(II) andMn(II), indicating that U, as well as Ba,may be involved in carrier-phase cycling processes.
4) The estuarine distribution of U in Tampa Bay iscontrolled by the underlying geology, as well as byenhanced fluid exchange processes across the sedi-
ment/water interface. While the pore waters or coas-tal aquifer underlying Tampa Bay mostly provide anet source of U to bay waters, site-specific U removalwas also observed. Such findings support the karsticnature of this estuary.
Acknowledgments
We thank Marci Marot (USGS), Jason Greenwood(ETI), Robyn Conmy (USF), and Sarah Trimble (WSU)for help in the field and in the laboratory. Alan Shiller(USM) provided expert high resolution ICP/MS traceelement analyses. Funding was graciously provided bythe Coastal Marine Geology Program of the USGSthrough the Tampa Bay Integrated Science Project. Wewould also like to thank Chuck Holmes (USGS) forhelpful early reviews, as well as WS Moore, the editor,and two anonymous reviewers, for their remarks whichgreatly improved the manuscript. The use of trade namesis for descriptive purposes only and does not imply en-dorsement by the U.S. Government.
References
Aller, R.C., Cochran, J.K., 1976. 234Th/238Udisequilibrium and diagenetictime scales. Earth and Planetary Science Letters 29, 37–50.
Aller, R.C., Mackin, J.E., Cox Jr., R.T., 1986. Diagenesis of Fe and Sin Amazon inner shelf muds: apparent dominance of Fe reductionand implications for the genesis of ironstones. Continental ShelfResearch 6, 263–289.
Aller, R.C., Blair, N.E., Xia, Q., Rude, P.D., 1996. Remineralizationrates, recycling, and storage of carbon in Amazon shelf sediments.Continental Shelf Research 16, 753–786.
Anderson, R.F., LeHuray, A.P., Fleisher, M.Q., Murray, J.W., 1989.Uranium deposition in Saanich Inlet sediments, Vancouver Island.Geochimica et Cosmochimica Acta 53, 2205–2213.
Andersson, P.S., Porcelli, D., Gustafsson, O., Ingri, J., Wasserburg, G.J.,2001. The importance of colloids for the behavior of uraniumisotopes in the low salinity zone of a stable estuary. Geochimica etCosmochimica Acta 65, 13–25.
Barnes, C.E., Cochran, J.K., 1990. Uranium removal in oceanicsediments and the oceanic U balance. Earth and Planetary ScienceLetters 97, 94–101.
Barnes, C.E., Cochran, J.K., 1993. Uranium geochemistry in estuarinesediments: controls on removal and release processes. Geochimicaet Cosmochimica Acta 57, 555–569.
Berner, R.A., 1980. Early Diagenesis: a Theoretical Approach.Princeton University Press, Princeton, N.J., 241 pp.
Bhat, S.G., Krishnaswami, S., 1969. Isotopes of uranium and radium inIndian rivers. Proceedings of the Indian Academy of Sciences 70,1–17.
Borole, D.V., Krishnaswami, S., Somayajulu, B.K.L., 1982. Uraniumisotopes in rivers, estuaries and adjacent coastal sediments ofwestern India: their weathering, transport and oceanic budget.Geochimica et Cosmochimica Acta 46, 125–137.
Boulad, A.P., Michard, G., 1976. Etuide de l’uranium, du thorium etde leurs isotopes dans quelques corottes du Basin Angolais
(Atlantique Sud-Est). Earth and Planetary Science Letters 32,77–83.
Boyle, E.A., Edmond, J.M., Sholkovitz, E.R., 1977. The mechanismof iron removal in estuaries. Geochimica et Cosmochimica Acta41, 1313–1324.
Brooks, G.R., Doyle, L.J., 1998. Recent sedimentary development ofTampa Bay, Florida: a microtidal estuary incised into Tertiaryplatform carbonates. Estuaries 21, 391–406.
Burnett, W.C., Bokuniewicz, H., Huettel, M., Moore, W.S., 2003.Groundwater and pore water inputs to the coastal zone. Bio-geochemistry 66, 3–33.
Carroll, J., Moore, W.S., 1994. Uranium removal during low dischargein the Ganges–Brahmaputra mixing zone. Geochimica et Cosmo-chimica Acta 58, 4987–4995.
Chen, J.H., Edwards, L.R., Wasserburg, G.J., 1986. 238U, 234U and 232Thin seawater. Earth and Planetary Science Letters 80, 241–251.
Choppin, R.C., Clark, S.B., 1991. The kinetic interactions of metalions with humic acids. Marine Chemistry 36, 27–38.
Church, T.M., Sarin, M.M., Fleisher, M.Q., Ferdelman, T.G., 1996.Salt marshes: an important coastal sink for dissolved uranium.Geochimica et Cosmochimica Acta 60, 3879–3887.
Cochran, J.K., 1982. The oceanic chemistry of the U- and Th-seriesnuclides. In: Ivanovich, M., Harmon, R.S. (Eds.), Uranium SeriesDisequilibrium: Applications to Environmental Problems. Clar-endon Press, Oxford, pp. 384–430.
Cochran, J.K., Carey, A.E., Sholkovitz, E.R., Suprenant, L.D., 1986.The geochemistry of uranium and thorium in coastal marinesediments and sediment pore waters. Geochimica et Cosmochi-mica Acta 50, 663–680.
Coffey, M., Dehairs, F., Collette, O., Luther, G., Church, T., Jickells,T., 1997. The behavior of dissolved Ba in estuaries. Estuarine andCoastal Shelf Science 485, 113–121.
Copenhaver, S.A., Krishnaswami, S., Turekian, K.K., Epler, N.,Cochran, J.K., 1993. Retardation of 238U and 232Th decay chainradionuclides in Long Island and Connecticut aquifers. Geochi-mica et Cosmochimica Acta 57, 597–603.
Davis, J.A., 1984. Complexation of trace metals by adsorbed naturalorganic matter. Geochimica et Cosmochimica Acta 48, 679–691.
Dooris, P.M., Dooris, G.M., 1985. Surface flows to Tampa Bay:quantity and quality aspects. Proceedings of Tampa Bay AreaScientific Information Symposium May 1982. University of SouthFlorida, Tampa.
Fleischer, R.L., Raabe, O.G., 1978. Recoiling alpha-emitting nuclei.Mechanisms for uranium series disequilibrium. Geochimica etCosmochimica Acta 42, 973–978.
Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N., Heath,G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B.,Maynard, V., 1979. Early oxidation of organic matter in pelagicsediments of the eastern equatorial Atlantic: suboxic diagenesis.Geochimica et Cosmochimica Acta 43, 1075–1091.
Gibbs, R.J., 1977. Transport phases of transitionmetals in theAmazon andYukon Rivers. Geological Society of America Bulletin 88, 829–843.
Green, R., Arthur, J.D., Dewitt, D., 1995. Lithostratigraphic andhydrostratigraphic cross-section through Pinellas and HillsboroughCounties, SW Florida. Florida Geologic Survey, Open-File Report61. 26 pp.
Hanor, J.S., Chan, L.H., 1977. Non-conservative behavior of Baduring the mixing of Mississippi River and Gulf of Mexico waters.Earth and Planetary Science Letters 43, 343–350.
Hsi, C.D., Langmuir, D., 1985. Adsorption of uranyl onto ferricoxyhydroxides: applications of the surface complexation site-bindingmodel. Geochimica et Cosmochimica Acta 49, 1931–1941.
Hussain, N., Krishnaswami, S., 1980. U-238 series radioactivedisequilibrium in groundwaters: implications to the origin of excessU-234 and fate of reactive pollutants. Geochimica et CosmochimicaActa 44, 1287–1291.
Klerks, P.L., Felder, D.L., Strasser, K., Swarzenski, P.W., 2007. Effectsof ghost shrimp on zinc and cadmium in Tampa Bay sediment.Marine Chemistry 104, 17–26.
Klinkhammer, G.P., Palmer, M.R., 1991. Uranium in the oceans:where it goes and why. Geochimica et Cosmochimica Acta 55,1799–1806.
Kolodny, Y., Kaplan, I.R., 1973. Deposition of uranium in the sedimentand interstitial water of an anoxic fjord. Proceedings of theSymposium on Hydrogeochemistry and Biogeochemistry: Hydro-geochemistry. Clark County, Washington, D.C, pp. 418–442.
Ku, T.L., Knauss, K.G., Mathieu, G.G., 1977. Uranium in open ocean:concentration and isotopic composition. Deep-Sea Research 24,1005–1017.
Langmuir, D., 1978. Uranium solution-mineral equilibria at lowtemperatures with application to sedimentary ore deposits.Geochimica et Cosmochimica Acta 42, 547–569.
Li, W.C., Victor, D.M., Chakrabarti, C.L., 1980. Effect of pH anduranium concentration on interaction of uranium (VI) and uranium(IV) with organic ligands in aqueous solutions. AnalyticalChemistry 52, 520–523.
Li, Y.H., Gregory, S., 1974. Diffusion of ions in seawater and deep-seasediments. Geochimica et Cosmochimica Acta 38, 703–714.
Li, Y.-H., Chan, L., 1979. Desorption of Ba and 226Ra from river-bornesediments in the Hudson estuary. Earth and Planetary ScienceLetters 43, 343–350.
Lienert, C., Short, S.A., von Gunten, H.R., 1995. Uranium infiltrationfrom a river to a shallow groundwater. Geochimica et Cosmochi-mica Acta 58, 5455–5463.
Mangini, A., Sonntag, C., Bertsch, G., Mueller, E., 1979. Evidence fora higher natural uranium content in world rivers. Nature 278,337–339.
Mann, D.K., Wong, G.T.F., 1993. “Strongly bound” uranium in marinewaters: occurrence and analytical implications. Marine Chemistry42, 25–37.
Martin, J.M., Hartl, K.M., Corbett, R.D., Swarzenski, P.W., Cable, J.E.,2003. A multi-level pore-water sampler for permeable sediments.Journal of Sedimentary Research 73, 128–132.
Martin, J.-M., Meybeck, M., 1979. Elemental mass-balance of materialcarried by major world rivers. Marine Chemistry 7, 173–206.
Martin, J.-M., Whitfield, M., 1983. The significance of the river input ofchemical elements to the ocean. In: Wong, C.S., Boyle, E., Bruland,K.W., Burton, J.D., Goldberg, E.D. (Eds.), TraceMetals in Seawater.NATO Conference Series. Plenum Press, New York. 265 pp.
McKee, B.A., DeMaster, D.J., Nittrouer, C.A., 1987. Uraniumgeochemistry on the Amazon shelf: evidence for uranium releasefrom bottom sediments. Geochimica et Cosmochimica Acta 51,2779–2786.
McManus, J., Berelson, W.M., Klinkhammer, G.P., Johnson, K.S.,Coale, K.H., Anderson, R.F., Kumar, N., Burdige, D.J., Hammond,D.E., Brumsack, H.J., McCorkle, D.C., Rushdi, A., 1998. Geochem-istry of barium in marine sediments: implications for its use as apaleoproxy. Geochimica et Cosmochimica Acta 62, 3453–3473.
Millward, G.E., Turner, A., 1994. Trace elements in estuaries. In:Salbu, B., Steinnes, E. (Eds.), Trace Elements in Natural Waters.CRC Press, Boca Raton, FL, pp. 223–246.
Moore, W.S., 1967. Amazon and Mississippi River concentrations ofuranium, thorium and radium isotopes. Earth and Planetary ScienceLetters 2, 231–234.
57P.W. Swarzenski, M. Baskaran / Marine Chemistry 104 (2007) 43–57
Moore, W.S., 1996. Large groundwater inputs to coastal watersrevealed by 226Ra enrichment. Nature 380, 612.
Moore, W.S., DeMaster, D.J., Smoak, J.M., McKee, B.A., Swarzenski,P.W., 1996. Radionuclide tracers of sediment–water interactionson the Amazon shelf. Continental Shelf Research 16, 645–665.
Nagao, S., Narita, H., Tsunogai, S., Harada, K., Ishi, T., 1992. Thegeochemistry of pore water uranium in coastal marine sedimentsfrom Funka Bay, Japan. Geochemistry Journal 26, 63–72.
Nozaki, Y., 1991. The systematics and kinetics of U/Th decay seriesnuclides in ocean water. Reviews in Aquatic Sciences 4, 75–105.
Osmond, J.K., Cowart, J.B., 1976. The theory and uses of naturaluranium isotopic variations in hydrology. Atomic Energy Reviews14, 621–679.
Osmond, J.K., Cowart, J.B., Humphreys, C.L., Wagner, B.E., 1984.Radioelement migration in natural and mined phosphate terrain.Florida Institute for Phosphate Research (FIPR) Report No. 05-002-027. 131 pp.
Palmer, M.R., Edmond, J.M., 1993. Uranium in river water. Geochimicaet Cosmochimica Acta 57, 4947–4955.
Payne, T.E., Waite, T.D., 1991. Surface complexation modeling ofuranium sorption data obtained by isotope exchange techniques.Radiochimica Acta 52/53, 487–493.
Plater, A.J., Ivanovich, M., Dugdale, R.E., 1992. Uranium seriesdisequilibrium in river sediments and waters: the significance ofanomalous activity ratios. Applied Geochemistry 7, 101–110.
Porcelli, D., Swarzenski, P.W., 2003. The behavior of U- and Th-seriesnuclides in groundwater and the tracing of groundwater. Reviewsin Mineralogy and Geochemistry 52, 317–361.
Presley, B.J., Trefry, J.H., Shokes, R.F., 1980. Heavy metal inputs toMississippi delta sediments. Water, Air, and Soil Pollution 13,481–494.
Rodushkin, I., Ruth, T., 1997. Determination of trace metals in estuarineand sea-water reference materials by high resolution inductivelycoupled plasma mass spectrometry. Journal of Analytical AtomicSpectrometry 12, 1181–1185.
Sackett, W.M., Cook, G., 1969. Uranium geochemistry of the Gulf ofMexico. Transactions of the Gulf Coast Association of theGeological Societies 19, 233–238.
Sarin, M.M., Church, T.M., 1994. Behavior of uranium during mixingin the Delaware and Chesapeake estuaries. Estuarine, Coastal andShelf Science 39, 619–631.
Scott, M.R., 1982. The chemistry of U and Th series nuclides in rivers.In: Ivanovich, M., Harmon, R.S. (Eds.), Uranium SeriesDisequilibrium: Applications to Environmental Problems. Clar-endon Press, Oxford, pp. 181–201.
Shaw, T.J., Sholkovitz, E.R., Klinkhammer, G., 1995. Redox dynamicsin the Chesapeake Bay: the effect of the sediment/water uraniumexchange. Geochimica et Cosmochimica Acta 58, 2985–2993.
Shaw, T.J., Moore, W.S., Kloepfer, J., Sochaski, M.A., 1998. The fluxof barium to the coastal waters of the southeastern USA: theimportance of submarine groundwater discharge. Geochimica etCosmochimica Acta 62, 3047–3054.
Shiller, A.M., Boyle, E.A., 1987. Variability of dissolved trace metalsin the Mississippi River. Geochimica et Cosmochimica Acta 51,3273–3277.
Shiller, A.M., Boyle, E.A., 1991. Trace elements in the MississippiRiver Delta outflow region: behavior at high discharge. Geochi-mica et Cosmochimica Acta 55, 3241–3251.
Shiller, A.M., Mao, L., 1999. Dissolved vanadium on the Louisianashelf: effect of oxygen depletion. Continental Shelf Research 19,1007–1020.
Sholkovitz, E., 1976. Flocculation of dissolved organic and inorganicmatter during the mixing of river water and seawater. Geochimicaet Cosmochimica Acta 40, 831–845.
Sholkovitz, E.R., 1977. The flocculation of dissolved Fe, Mn, Al, Cu,Ni, Co, Cd during the estuarine mixing. Earth and PlanetaryScience Letters 41, 77–86.
Sholkovitz, E.R., Boyle, E.R., Price, N.B., 1978. Removal of dissolvedhumic acid and iron during estuarine mixing. Earth and PlanetaryScience Letters 40, 13–136.
Snow, D.D., Spalding, R.F., 1994. Uranium isotopes in the Platte Riverdrainage basin of the North American High Plains Region. AppliedGeochemistry 9, 271–278.
Somayajulu, B.L.K., 1994. Uranium isotopes in the Hooghly estuary.Marine Chemistry 47, 291–296.
Swarzenski, P.W., McKee, B.A., 1999. Seasonal uranium distributionin the coastal waters off the Amazon and Mississippi Rivers.Estuaries 21, 379–390.
Swarzenski, P.W., McKee, B.A., Booth, J.G., 1995. Uranium geochem-istry on the Amazon Shelf: chemical phase partitioning and cyclingacross a salinity gradient. Geochimica et Cosmochimica Acta 59,7–18.
Swarzenski, P.W., McKee, B.A., Skei, J.M., Booth, J.G., Todd, J.F., 1999.Uranium across the redox transition zone of a permanently stratifiedFjord: Framvaren, Norway. Marine Chemistry 67, 181–198.
Swarzenski, P.W., Reich, C.D., Spechler, R.M., Kindinger, J.L.,Moore, W.S., 2001. Using multiple geochemical tracers tocharacterize the hydrogeology of the submarine spring off CrescentBeach, Florida. Chemical Geology 179, 187–202.
Swarzenski, P.W., Porcelli, D., Andersson, P., Smoak, D., 2003. Thebehaviour of U- and Th-series nuclides in the estuarine environ-ment. Reviews in Mineralogy and Geochemistry 52, 577–606.
Swarzenski, P.W., Campbell, P.L., Porcelli, D., McKee, B.A., 2004. Theestuarine chemistry and isotope systematics of 234,238U in theAmazon and Fly Rivers. Continental Shelf Research 24, 2357–2372.
Swarzenski, P.W., Reich, C., Kroeger, K.D., Baskaran, M., 2007. Raand Rn isotopes as natural tracers of submarine groundwaterdischarge in Tampa Bay, FL. Marine Chemistry 104, 69–84.
Trefry, J.H., Nelson, T.A., Trocine, R.P., Metz, S., Vetter, T.W., 1986.Trace metal fluxes through the Mississippi River Delta system.Rapports et proces-verbaux des Reunions - Conseil Internationalpour l'Exploration de la mer 186, 277–288.
Waite, T.D., Davis, J.A., Payne, T.E., Waychunas, G.A., Xu, N., 1994.Uranium(VI) adsorption to ferrihydrite: application of a surfacecomplexation model. Geochimica et Cosmochimica Acta 58,5465–5478.
Weisberg, R.H., Zheng, L., 2006. The circulation of Tampa Bay drivenby buoyancy, tides, and winds, as simulated using a finite volumecoastal ocean model. Journal of Geophysical Research 111,C01005, doi:10.1029/2005JC003067.
Zielinski, R.A., Meier, A.L., 1988. The association of uranium withorganic matter in Holocene peat: an experimental leaching study.Applied Geochemistry 3, 631–643.
Zhorov, V.A., Boguslavskiy, S.G., Kir'yakov, P.A., 1982. Geochem-istry of uranium in the Black Sea. Geochemistry International 19,191–200.
