Estuaries and Coasts DOI 10.1007/s12237-013-9719-8

Metabolic Responses of Estuarine Microbial Communities to Discharge of Surface Runoff and Groundwater from Contrasting Landscapes Patrick R. Hutchins & Erik M. Smith & Eric T. Koepfler & Richard F. Viso & Richard N. Peterson

Received: 12 December 2012 / Revised: 3 September 2013 / Accepted: 26 September 2013 # Coastal and Estuarine Research Federation 2013

Abstract Groundwater discharge is increasingly recognized as a significant source of nutrient input to coastal waters, relative to surface water inputs. There remains limited information, however, on the extent to which nutrients and organic matter from each of these two flowpaths influence the functional responses of coastal microbial communities. As such, this study determined dissolved organic carbon (DOC) and nutrient concentrations of surface water runoff and groundwater from both an urbanized and a relatively pristine forested drainage basin near Myrtle Beach, South Carolina, and quantified the changes in production rates and biomass of phytoplankton and bacterioplankton in response to these inputs during two microcosm incubation experiments (August and October, 2011). Rainwater in the urbanized basin that would otherwise enter the groundwater appeared to be largely rerouted into the surface flowpath by impervious surfaces, bypassing ecosystem buffers and filtration mechanisms. Surface runoff from the developed basin was most enriched in nutrients and DOC and yielded the highest production rates of

Communicated by James L. Pinckney P. R. Hutchins : R. F. Viso : R. N. Peterson Center for Marine and Wetland Studies, School of Coastal Marine and Wetland Systems Science, Coastal Carolina University, Conway, SC 29526, USA P. R. Hutchins (*) Division of biological Sciences, University of Montana, Missoula, MT 59801, USA e-mail: [email protected] E. M. Smith North Inlet-Winyah Bay National Estuarine Research Reserve, and the Baruch Marine Field Laboratory, University of South Carolina, Georgetown, SC 29442, USA E. T. Koepfler Department of Marine Science, Coastal Carolina University, Conway, SC 29526, USA

the various source waters upon addition to coastal waters. The metabolic responses of phytoplankton and bacterioplankton were generally well predicted as a function of initial chemical composition of the various source waters, though more so with bacterial production. Primary and bacterial productivities often correlated at reciprocal time points (24-h measurement of one with the 72-h measurement of the other). These results suggest human modification of coastal watersheds enhances the magnitude of dissolved constituents delivered to coastal waters as well as alters their distributions between surface and groundwater flowpaths, with significant implications for microbial community structure and function in coastal receiving waters. Keywords Phytoplankton . Bacteria . Production . Stormwater . Groundwater . Estuaries

Introduction Urbanization is one of the most pervasive features of global environmental change and impacts ecosystem processes from local to global scales (Vitousek et al. 1997; Grimm et al. 2008). The conversion from forest and agricultural lands to urban land uses profoundly impacts watershed hydrology as well as the quantity and forms of terrestrial materials and pollutants introduced to downstream receiving waters (Walsh et al. 2005). This is especially true in the coastal zone, where much of the population growth and development is occurring (Crossett et al. 2004; Agardy and Alder 2005). Coastal development represents a particularly pressing issue for the Southeastern region of the USA, which is experiencing some of the most rapid population growth in the nation and where land development rates are often much greater than the rate of population growth (Allen and Lu 2003). As is the case in other regions, stormwater runoff associated with urbanization in the Southeastern coastal plain has been

Estuaries and Coasts

shown to contain higher concentrations of nutrients, suspended solids, organic contaminants and fecal coliforms, relative to runoff from more forested watersheds (Tufford et al. 2003; DiDonato et al. 2009; Mallin et al. 2009). Although dissolved organic carbon (DOC) concentrations tend to be lower in urban runoff relative to runoff from forested lands (Wahl et al. 1997), the dissolved organic material carried in urban runoff tends to be more bioreactive than that in forest runoff (Mallin et al. 2009). As a result of these alterations in material concentrations, degradation of estuarine tidal creeks and larger coastal waters has been positively linked to the proportion of impervious surface associated with development in coastal watersheds within the Southeast (Sanger et al. 1999; Mallin et al. 2000; Holland et al. 2004; Van Dolah et al. 2008). While material discharge in urban runoff can directly affect water quality, urban development may also indirectly affect water quality by rerouting natural water flowpaths, bypassing soil and vegetation buffers that provide extensive ecosystem services (Woodward and Wui 2001). Impervious surfaces and stormwater infrastructure, for instance, concentrate and constrain rainwater to limited points of flow, thus evading chemical and biological transformations that would otherwise occur through infiltration prior to subsurface discharge to receiving waters (Walsh et al. 2005). In addition, extensive impervious surface coverage can alter the partitioning of water between surface and sub-surface flowpaths (Paul and Meyer 2001). This may be especially relevant in the Southeastern coastal plain, where surface and sub-surface flowpaths are naturally intimately linked due to the region's low topographic relief, generally sandy soils, and shallow water table (Harder et al. 2007; Epps et al. 2012). Due to the physical, chemical and biological processes occurring in soils, concentrations of nutrients in groundwaters can be very different, and exist in very different relative proportions, relative to those of corresponding surface runoff (Slomp and Van Cappellen 2004). These differences can have consequences for the biological dynamics of coastal receiving waters (e.g., Lapointe et al. 1999). This is not a trivial issue, given that it has been shown that total nutrient fluxes associated with coastal groundwater discharge can often rival that of nutrient fluxes associated with riverine discharge (Moore 1999; Santos et al. 2010; Swarzenski et al. 2007). Prior studies of groundwater geochemical composition have focused on inorganic nutrients, with little information available regarding organic nutrient forms or DOC in coastal groundwater and how they compare to that of surface water runoff in terms of concentration, composition and bioavailability. In addition, few studies have directly quantified the effects of groundwater discharge on coastal marine microbial communities under controlled conditions (Garcès et al. 2011). At present, there exists rather limited information to characterize the relative balance of carbon and nutrient concentrations in both surface stormwater runoff and groundwater input at locations that contrast in their degree of urbanization, and,

to our knowledge, no studies that directly compare the effects of groundwater and surface water on the metabolic responses of coastal microbial communities. Such information is necessary, however, in order to understand how continued rapid urbanization of coastal watersheds, especially in the Southeast USA, will affect natural environmental controls over microbial community structure and function within coastal marine ecosystems. The aim of the present study is to determine the role of land use (urbanized vs. forested) and flowpath (surface vs. subsurface) on the chemical composition of discharging waters and resulting biological responses of estuarine microbial communities to these different input sources. To do so, we contrast surface runoff and groundwaters from Myrtle Beach, South Carolina, to those of a nearby forested drainage basin that exists in its relatively natural and pristine state. The greater Myrtle Beach area is a highly urbanized beachfront resort community. It had a 2010 resident population of ~300,000 and was the nine fastest-growing metropolitan area in the US in 2010 (US Census Bureau). In addition, it receives more than 14 million tourist visitors annually (Myrtle Beach Chamber of Commerce), with peak visitor numbers occurring in the summer months. We focus on microbial community responses to varying source water inputs because microorganisms are strongly influenced by both allochthonous terrestrial material delivery and internal cycling of nutrients and organic matter and because they are capable of quickly altering their metabolic rates and abundances in response to environmental perturbations such as anthropogenic nutrient and organic matter loading (e.g., Paerl et al. 2003). Furthermore, they form the base of the food web in most marine ecosystems and dominate ecosystem scale material and energy fluxes (Azam et al. 1982). Thus, quantifying how microbial communities and their metabolic rates are affected by shifts in the magnitude and route of material delivery from terrestrial to aquatic ecosystems is a key component of understanding how coastal development impacts ecosystem functioning and eutrophication processes in marine receiving waters.

Methods The goal of the experimental design was to assess the ecological impacts of surface runoff and groundwater, from both a forested and an urban drainage basin (Fig. 1), when discharged into coastal marine waters. The experimental approach was similar to commonly employed enrichment bioassay experiments designed to assess the effects of nutrient or organic amendment on microbial communities (e.g., Hitchcock et al. 2010; Siegel et al. 2011), except that natural aliquots of surface and groundwater sources were used, rather than defined artificial substrates. In doing so, the experiment attempts to simulate the pulsed input of materials from each of

Estuaries and Coasts

a

North Carolina

Source Water Collection Sites DGw DS w

South Carolina

Horry County

PGw PS w

C

NIw Drainage Basin

ng

Ba y

Georgetown County

Lo

B

0

12 .5 2 5

50 km

b

c

k

ree

lC hau

b

Cra 0

0. 5

1

0

2

km

0. 25

0 .5

1

km

Fig. 1 The northern South Carolina coastal region (a) and sample sites and the surface runoff catchments (outlined areas) for source waters from our pristine site (b) and developed site (c). PGw pristine groundwater, PSw pristine surface runoff, DQw developed groundwater, DSw developed surface runoff

the source waters as a result of a single rain-induced discharge event. This was accomplished by adding equivalent filtered volumes of each source water in separate treatments to an unaltered water sample from a relatively pristine coastal estuary and quantifying phytoplankton and bacterioplankton production rates and abundances over 5 days of incubation, a period of time that would allow for community-level phytoplankton responses to enrichment (Downing et al. 1999). Site Description and Sample Collection To represent pristine groundwater (PGw) and pristine surface water (PSw), we sampled the Oyster Creek drainage (Fig. 1b), which exists in a natural forested state and typifies a forested drainage basin of the Southeastern coastal plain (Wahl et al. 1997). Thus, the site is pristine to the extent that it exists in an

undeveloped, naturally forested state. Oyster Creek experiences tidal inundation at its lower extent and is dry at higher elevations, except during periods of rainfall. PGw samples were obtained using a peristaltic pump and four PVC piezometers spaced 15 m apart and inserted 1.5 m below the sediment surface at the marsh interface, between pine forest and Juncus salt marsh grass zones (33°21′16″ N, 79°11′38″ W). Groundwater sampling sites were located such that we could collect the most biologically relevant source water (i.e., nearest the point of discharge) without concerns regarding the compounding effects of recirculated estuarine or ocean water. In this way, we hoped to measure and test "new" meteoric water that was being added to the system. PSw was collected from surface runoff in the upper reaches of Oyster Creek (33°21′32″ N, 79°11′37″ W), where a catchment concentrates storm flow for easy surface sample collection.

Estuaries and Coasts

To represent groundwater and surface water runoff from a highly developed coastal drainage (DGw and DSw, respectively), we selected a drainage located in the heavily developed area of Myrtle Beach, SC (Fig. 1c). A highly engineered system of ditches, culverts, and pipes drains stormwater from the basin to Withers Swash, a tidal creek that discharges to the Atlantic Ocean across the beachface. Withers Swash is tidally inundated, with the tidal wedge propagating about 1.7 km upstream at spring tide, thus offering a developed analog to the pristine Crabhaul Creek sub-estuary of North Inlet. DSw was collected from the outlet of a stormwater discharge pipe (33°41′32″ N, 78°53′32″ W). DGw was collected from 1.5 m depth using a peristaltic pump and PVC piezometers located in the same drainage network and just above the high tide line prior to draining to Withers Swash (33°41′30″ N, 78°53′35″ W). ESRI ARC-GIS was used to determine land cover classes of each drainage basin using data from the National Land Cover Institute and USGS Lidar data from the national elevation dataset. The area draining the pristine study site was approximately 0.37 km2 and covered by 85 % wooded and herbaceous wetland and 15 % evergreen forest. The 0.34-km2 drainage area at the developed site is comprised of 59 % commercial, 37 % residential, and 4 % other land use classifications and is 46 % impervious cover. Both sampling sites only experience surface runoff in response to rain events, with no permanent base flow, such that flow from these basins is roughly proportional to rain fall amount, with the exact relationship depending on season and antecedent soil moisture conditions (e.g., Epps et al. 2012). Surface and groundwater samples from both pristine and developed locations were collected after rain events occurring on two occasions, once in the summer (July 25) and once in autumn (October 10). The July sampling represents the period in which Myrtle Beach tourist populations are at their peak, as well as the time that water temperatures and microbial activity are at their annual maxima in coastal waters of this area. The October sampling represents a period of substantially subsiding tourist activity as well as the waning of biological activity in coastal waters. Total precipitation for the July event was approximately 43 mm and approximately 20 mm for the October event. Both these rainfall totals fall within the second quartile (12.4–66.7 mm) of long-term rainfall totals for this region (unpublished data from the North Inlet meteorological station). The last significant rain event prior to the July 25th sampling occurred on July 9, and the last significant rain event prior to October 10th sampling occurred on September 30. Upon collection, all source water samples were GF/F filtered (nominal pore size=0.7 μm), sub-sampled for dissolved nutrient and organic carbon analyses (described below), and frozen (−80 °C) until the start of the mixing experiments. The lower reaches of the Crabhaul Creek sub-basin of North Inlet estuary, located near Georgetown, SC (Fig. 1), was chosen to represent the coastal marine receiving waters

into which terrestrial source waters were mixed. Crabhaul Creek is a well-mixed, tidally influenced, ocean-dominated salt-marsh creek (mean tidal range is ~1.5 m, annual mean salinity is ~32; Kjerfve 1986) in the relatively pristine North Inlet estuary. It is located at the northwestern extent of approximately 34 km2 of Holocene marsh and is confined to the west by low-lying upland pine forest and is the receiving body of water for discharges from the Oyster Creek sampling site (Fig. 1b). Unfiltered samples of this tidal creek water (referred to as NIw) containing an intact native microbial assemblage were collected at the Oyster Landing monitoring station of the North Inlet–Winyah Bay National Estuarine Research Reserve (NIWB NERR) in North Inlet (33°20′02″ N, 79°11′34″ W) immediately prior to the start of the two mixing experiments. Experimental Incubation Set-up Mixing experiments began on August 8 and October 10, 2011. Mean water temperatures for the incubations were 30 °C and 21 °C, respectively. Light and precipitation levels for each incubation are shown in Fig. 2. To initiate each experiment, replicate (n =3) aliquots of 400 ml of each of the four source waters (PGw, PSw, DGw and DSw) plus an ultra-pure deionized control water (Cw) were mixed with 1,600 ml of the NIw receiving water in 2-l clear polycarbonate bottles. We elected for a constant 20 % source water treatment, as ongoing studies (Peterson, unpublished data) have shown that the discharge waters of Withers Swash contain an average of roughly 20 % groundwater. Samples were incubated under a shade cloth (allowing~60 % ambient surface light) in an outdoor flowthrough incubation system cycling water from Crabhaul Creek, thus maintaining in situ light/dark cycles and temperatures. Sub-samples for autotrophic and heterotrophic production rates and abundances, as well as nutrient and organic carbon analyses were taken immediately upon mixing and again at 24, 72 and 120 h after the start of the incubation, using acid-cleaned HDPE bottles. Sub-samples for nutrient and organic carbon analyses were filtered through GF/F filters and either refrigerated at 4 °C (for those samples measured on the same day) or frozen (−80 °C) until analysis. Analytical Methods Total dissolved nitrogen (TDN) was measured using a Shimadzu TNM-1 autoanalyzer according to standard methods (4500-N; Eaton and Franson 2005). Dissolved nitrate (NO3−) plus nitrite (NO2−) concentrations (referred to as NOx ) were determined using triplicate 100-μl injections in an Antek 7000 element analyzer with an Antek 745 nitrate/nitrite attachment (Garside 1982). A modified version of the phenylhypochlorite method of Solorzano (1969) was used to determine dissolved ammonium (NH4+) concentrations immediately after sample collection. The combined quantity of dissolved nitrate, nitrite,

Estuaries and Coasts Fig. 2 Cumulative daily precipitation and photosynthetically active radiation (PAR) recorded at North Inlet for the days leading up to and during incubation experiments carried out in August (a) and October (b)

and ammonium is hence forth referred to as dissolved inorganic nitrogen (DIN). Total dissolved phosphorous (TDP) was measured spectrophotometrically (Koroleff 1983) following the high temperature combustion method of Solorzano and Sharp (1980). Soluble reactive phosphorous (SRP) was determined immediately after sample collection spectrophotometrically using the method of Koroleff (1983). Dissolved organic phosphorus and nitrogen (DOP and DON) were estimated by subtracting SRP from TDP and DIN from TDN, respectively. Filtrate aliquots were immediately acidified to pH <2 with 10 % HCL and stored in the dark at 4 °C until analyzed for DOC via a Shimadzu TOC-VCPN organic carbon analyzer following Benner and Strom (1993). Phytoplankton abundance was assessed as chlorophyll a (Chl) concentration. Samples for Chl analyses were collected on GF/F filters, extracted in acetone (90 % for 48 h at 4 °C) and quantified fluorometrically on a Turner Designs fluorometer following EPA standard method 445.0 (Arar and Collins 1997).

Bacterial abundance (BA) was determined by staining with Acradine Orange (200 μg/ml), filtering onto 0.4-μm black PCTE membrane filters (Millipore) and quantifying with epifluorescent microscopy at 1,000× magnification (Daley and Brown 1973). The first field of view was counted in its entirety or until reaching ~40 cells, whichever occurred first. Multiple fields of view were counted until either 400 cells had been counted or until 20 fields had been observed, whichever occurred first. No less than nine fields of view were counted for any sample (Kirchman et al. 1982). Phytoplankton primary production rates (PP) were quantified on time-course sub-samples using a modification of the 14 C bicarbonate incorporation followed by acidification and bubbling method of Schindler et al. (1972). Sub-samples were spiked with NaH14CO3 for a final activity of 1 μCi ml−1 and incubated for 1 h at average in situ temperatures for each mixing experiment and an irradiance of 210 μmol photons m−2 s−1, which is close to the previously reported annual mean

Estuaries and Coasts

value for the photosynthesis light saturation value (E k ) for North Inlet (Lawrenz et al. 2013). Incubations were terminated by addition of 250 μl buffered formalin (5 % final concentration), acidified and shaken for 12 h to degas unincorporated 14 C. Activity of 14C was then counted for 5 min on a PerkinElmer Tri-Carb 2100 liquid scintillation counter using Ecolume (MP Biomedicals) as a scintillation cocktail. Disintegrations per minute (dpm) were converted to carbon incorporation rates (PP) according to Knap et al. (1996) with isotope dilution determined by quantifying dissolved inorganic carbon concentration on acidified aliquots of sample using a LI-COR LI 7000 infrared CO2 gas analyzer. Bacterial production rates were determined from incorporation rates of 3H leucine into the cold trichloroacetic acid (TCA; 5 % final concentration)-insoluble fraction of macromolecules (1.5 ml, three replicates plus one killed control) incubated with 35 nM (final concentration) of 3H leucine (~5 TBq mmol−1) at in situ temperatures for 1 h in the dark. TCA precipitated macromolecules were collected following the centrifugation method of Smith and Azam (1992) and 3H activity was counted for 3 min on a PerkinElmer Tri-Carb 2100 liquid scintillation counter using Ultima Gold (PerkinElmer) as a scintillation cocktail. Incorporation rates of 3H leucine were converted to carbon production rates with a conversion factor of 3.1 kg carbon (mol leucine)−1 (Kirchman 1993). Statistical Analysis One-way analysis of variance (ANOVA) with Tukey's post-hoc test at the 95 % confidence limit (a =0.05) was used to compare source water concentrations. In an effort to conservatively determine confidence intervals and test for significant differences between treatments, only bottles within a treatment were considered as replicates (n =3) for ANOVAs, regardless of additional replication of chemical and bioassay measurements. In instances with analytical replication, a mean of the measured parameter was considered representative of the bottle. All data were subjected to a Shapiro–Wilk normality test and Levene's test of equality of variance prior to analysis. Violations of these assumptions were resolved by log transforming data. Repeated measures analysis of variance (RMANOVA) tests were performed for samples pooled across all treatments within each incubation (n =15) to identify significant effects of basin (pristine or developed, between-subject factor), flowpath (groundwater or surface runoff, between-subject factor), treatment (PGw, PSw, DGw, and DSw, betweensubject factor) and sample time (0, 24, and 72 h; withinsubject factor) on biological response variables (PP,BP, Chl, and BA). We elected not to use the sample points at 120 h due to the unfortunate lack of PP data (a key response variable) at this time point in October experiments (due to logistical issues). Non-normally distributed data were log-transformed prior to analysis. In cases where the assumption of sphericity

was violated, a Huynh–Feldt correction was used to adjust p values (Scheiner and Gurevitch 2001). All statistical tests were performed using IBM SPSS v20 statistical software.

Results Composition of Source Waters Among source waters collected in July, DSw had the highest concentrations of total, organic, and inorganic nutrients as well as DOC (see Table 1 for statistically significant groups). DGw had the lowest nutrient and organic matter concentrations with the exception of DOP and SRP. PGw had higher TDN than PSw, but PSw was dominated by inorganic nitrogen, in the form of NOx , while PGw nitrogen was mostly DON and NH4. SRP concentrations were nearly identical between PGw and PSw, which had the lowest SRP and DOP concentrations among the four source water types. While surface water DOC concentrations were higher than those of groundwater in the developed basin, the opposite holds true in the pristine site. Source waters collected from the pristine site were slightly saline (salinity was 2.2 for PSw and 5.9 for PGw), whereas source waters from the developed site were essentially freshwater (salinities <0.5 for both DSw and DGw). For samples collected in October, DSw again had the highest concentrations of inorganic and organic nitrogen and phosphorus but had slightly less DOC than PGw. Similar to August, SRP from DGw was higher than PGw and PSw, but DGw was lowest in all other parameters measured. PGw had much higher DIN (NH4 +NOx ) and DON than PSw but only slightly more SRP. As expected for reducing groundwaters, PGw DIN was nearly all in the form of NH4. DOC was highest in PGw, which was slightly more concentrated than DSw. As was the case in July, source waters from the developed site were essentially freshwater (salinities <0.5) while pristine waters were slightly saline (salinity was 2.5 for PSw and 6.5 for PGw). Microcosm Incubations Initial incubation water nutrient and organic carbon concentrations (after amending NIw with 20 % source waters) are summarized in Table 2. In general, DSw had the highest concentration of nutrient and organic analytes while DGw and Cw bottles had the lowest. No significant difference existed in the Chl concentration between treatments at T 0 in August or October (p >0.1), thus any significant differences in autotrophic productivity or abundance between treatments should not be an artifact of unequal cell distribution in incubation bottles. The average initial Chl concentration in August incubation bottles was 5.71±0.07 μg l−1, somewhat higher than October (4.56±0.09 μg l−1). For August, none of the initial BA measurements were significantly different (mean

Estuaries and Coasts Table 1 Nutrient and organic carbon concentrations (μM) in source waters

August

October

NH4−a NO2− +NO3−a TDNa SRPa TDPa DOC NH4− NO2− +NO3− TDN SRP TDP DOCb

PGw

PSw

DGw

DSw

41.25±0.39 c 0.63±0.13 b 114.67±2.87 c 1.00±0.10 a 2.26±0.53 b 3044.10±67.14 c 44.36±1.45 b 0.06±0.10 a 168.56±2.73 c 1.45±0.04 b 1.21±0.20 ab 3020.56±0.00 c

4.88±1.37 b 74.12±12.05 d 100.86±4.95 b 1.10±0.01 bz 1.64±0.14 a 1207.24±50.72 b 3.08±0.44 a 2.32±0.09 c 36.67±0.01 b 0.74±0.06 a 0.83±0.39 a 1636.49±0.00 b

3.13±1.59 a 0.22±0.06 a 9.44±0.92 a 2.25±0.13 c 2.11±0.59 b 161.85±37.42 a 2.82±0.59 a 0.27±0.06 b 9.19±3.56 a 2.22±0.12 c 1.99±1.81 b 176.64±0.00 a

47.15±1.38 c 58.37±0.70 c 487.48±7.60 d 13.13±0.29 d 10.98±2.11 c 9652.14±629.41 d 98.21±5.73 c 105.04±5.72 d 366.22±8.64 d 8.81±0.14 d 12.76±5.36 c 2618.42±0.00 d

Letters denote significantly different groups within treatments for each parameter (one-way ANOVA; p <0.05, post-hoc Tukey's test). All data logtransformed prior to ANOVA a

Normality or equal variance test not passed after log transformation

b

October DOC measurements were not replicated

9.87±0.74×105 cells l−1; p >0.5). In October, only initial DSw BA was significantly higher than Cw, but was not significantly different from the other treatments. Initial BA for the October incubation averaged 4.70±0.51×105 cells l−1.

Once treatments were added to NIw, sharp decreases in inorganic nutrient concentrations occurred within 24 h in the August incubation (Fig. 3). SRP and NH4 were almost entirely consumed in all treatment bottles within 24 h and NOx was

Table 2 Nutrient and organic carbon concentrations (μM) and microorganism abundance in incubation microcosms at T 0

August

October

Cw

PGw

PSw

DGw

NH4−a NO2− +NO3−a TDN SRP TDP DOC TDN/TDP DIN/DIP Chl a (μg Chl ml−1) BA (105 cells ml−1) NH4−a,b NO2− +NO3−a TDN SRP TDP

3.21±0.80 ab 0.62±0.05 a 16.57±1.28 a 0.81±0.29 bz 1.63±0.41 a 149.72±4.41 a 11.69±2.42 a 8.12±5.09 a 5.56±0.24 a 10.98±2.81 a 3.11±0.20 b 0.90±0.05 b 21.06±1.08 a 0.21±0.03 a 0.55±0.11 a

3.42±0.22 ab 0.63±0.03 a 31.26±0.41 b 0.11±0.03 a 0.66±0.07 a 589.11±21.98 c 48.62±5.36 b 41.20±12.63 b 5.74±0.19 a 11.22±1.77 a 11.71±0.14 c 0.51±0.12 a 52.15±1.23 b 0.22±0.02 a 0.64±0.23 a

2.25±0.15 a 11.54±0.07 b 32.60±0.65 b 0.35±0.03 abz 1.34±0.30 a 361.07±8.80 b 26.93±9.17 a 39.14±2.99 b 5.86±0.09 a 10.58±0.89 a 2.50±0.33 a 1.25±0.12 b 28.51±0.71 a 0.26±0.04 ab 0.38±0.07 a

2.42±0.11 a 0.50±0.01 a 13.25±0.25 a 0.27±0.01 ab 0.82±0.02 a 165.20±2.70 a 13.61±0.36 a 9.51±1.07 a 5.65±0.12 a 8.73±1.46 a 2.47±0.08 a 0.55±0.05 a 22.38±1.30 a 0.37±0.04 b 0.57±0.08 a

4.58±0.14 b 18.60±0.51 c 111.11±0.74 c 2.41±0.04 cz 4.41±0.05 b 1968.15±4.54 d 25.22±0.33 a 9.52±0.36 a 5.71±0.18 a 7.85±0.88 a 12.65±0.05 c 30.87±0.78 c 103.55±3.88 c 1.75±0.02 cz 4.42±1.59 b

DOC TDN/TDP DIN/DIP Chl a (μg Chl ml−1) BA (105 cells ml−1)

128.99±0.86 a 40.32±5.34 a 19.32±2.48 b 4.42±0.16 a 3.49±1.30 ab

671.52±4.25 e 104.07±31.40 a 56.20±7.42 c 4.50±0.23 a 2.13±0.19 a

449.90±0.89 c 81.21±15.29 a 15.09±2.15 b 4.81±0.17 a 5.04±0.70 ab

167.67±3.61 b 41.75±8.95 a 8.41±0.90 a 4.63±0.07 a 0005.03±0.72 ab

633.17±1.90 d 29.33±7.93 a 24.88±0.47 b 4.45±0.39 a 6.96±0.58 b

Letters denote significantly different groups within treatments for each parameter (one-way ANOVA; p <0.05, post-hoc Tukey's test) a

Data log-transformed prior to ANOVA

b

Normality or equal variance test not passed after log transformation

DSw

Estuaries and Coasts

PGw

Cw

PSw

5

DSw

DGw

August

October

a

b

15

(µM)

4

NH

4

Fig. 3 Changes in the concentrations of NH4 (a and b), NOx (c and d), SRP (e and f), and DOC (g and h) over 5-day incubations in August (left) and October (right). Note the differences in y-axis scales between August and October

10

3 2

5

1 0

0

c

NO x (µM)

20 15

30

10

20

5

10

0

0

e

SRP (µM)

3

2

1

1

0

0

g

DOC (µM)

f

3

2

2000

d

40

h

1000 800

1500

600 1000 400 500

200

0

0 0

24

72

120

0

24

72

120

Incubation Time (hrs)

entirely consumed within 72 h. Nutrients were consumed much more slowly in the October incubation. All inorganic nutrients had been consumed after 72 h, with the exceptions of NH4 (PGw) and NOx (DSw) which were still present at the end of the 120-h incubation. Among all incubations, only three net decreases in DOC concentration occurred, indicating an organic carbon demand greater than that supplied by in situ processes. In August, only the DSw bottles exhibited net DOC

consumption (Fig. 3g). In October, both the DSw and PGw bottles showed net consumption of DOC in the first 24 h (Fig. 3h). Patterns in PP at 24 h agreed well with those of abundance during both experiments (Fig. 4). Surface water treatments generally produced significantly greater increases in PP and Chl than groundwater treatments (Table 3). Chl was only stimulated by surface water treatments in both experiments

Estuaries and Coasts

PGw

Cw

PSw

(µg mL-1)

50

October

30

60

20

40

10

20

0

0 15

c

20

d

10

15 10

5

5 0

120

0 100

e

(µg CL-1 hr-1)

80

60

60

40

40

20

20

(µg CL-1hr-1)

25

f

80

100

0

0

Bacterial Productivity

b

25 (105 Cells mL-1)

Bacterial Abundance

100 80

140 Primary Productivity

a

40

30

DSw

DGw

August

Ch l

Fig. 4 Chlorophyll a concentrations (a and b), bacterial abundances (c and d), primary production (e and f), and bacterial production (g and h) over 5-day incubations in August (left) and October (right). Note the differences in y-axis scales between August and October

10

g

20

8

15

6

10

4

5

2

0

0 0

24

72

120

h

0

24

72

120

Incubation Time (hrs)

(Table 3). PP in August was stimulated by all treatments (except DGw) relative to the control. Peaks in PP and Chl typically occurred at 24 h in the August incubation. In October, only the surface water treatments resulted in 24-h increases in Chl, and only DSw continued to increase through 72 h, although these levels crashed by 120 h. In the October incubation, peak Chl response from DSw in was more than twice the peak response in August. All PP increased after 24 h in October except for PGw. Though PGw PP increased throughout the experiment, its net effect (as with PSw) was inhibitory relative to the control

treatment (Table 3). DSw alone showed a relative stimulation of PP. Only one of three DSw treatment replicates was successfully analyzed for PP at 72 h. However, variation was extremely low between replicates within the DSw treatment at both prior times (~2 % error), suggesting that the one DSw bottle measured at 72 h is likely a true response. DSw was the only treatment that produced significantly increased BA after 24 h in the August incubation (Fig. 4). BP in the August DSw treatments increased rapidly in the first 24 h, peaked at 72 h, and remained high at 120 h. PGw yielded

Estuaries and Coasts Table 3 The difference of means from the control (treatment−control), significance, and net effect of treatments for biological response variables chlorophyll a concentration (Chl), bacterial abundance (BA), primary production (PP), and bacterial production (BP) as determined by RMANOVA with Bonferroni's post-hoc test (n =30)

Variable

Treatment

August ΔMeans

Chl (μg Chl ml−1) BA (105 cells ml−1) PP (μg C l−1 h−1) BP (μg C l−1 h−1)

October p

Effect

ΔMeans

p

Effect

PGw

4.55

0.50

Not significant

1.71

1.00

Not significant

PSw DGw DSw PGw PSw DGw DSw PGw PSw DGw DSw PGw PSw DGw DSw

12.51 0.25 17.93 −0.89 0.70 −1.25 6.20 14.40 38.04 −0.14 36.72 3.30 8.33 0.96 11.52

<0.01 1.00 <0.01 1.00 1.00 1.00 0.02 <0.01 <0.01 1.00 <0.01 <0.01 <0.01 1.00 <0.01

Stimulatory Not significant Stimulatory Not significant Not significant Not significant Stimulatory Stimulatory Stimulatory Not significant Stimulatory Stimulatory Stimulatory Not significant Stimulatory

7.60 0.63 35.87 0.91 3.22 2.45 5.74 −17.95 −9.31 −2.73 21.81 2.04 0.41 0.35 3.11

0.03 1.00 <0.01 1.00 1.00 1.00 0.11 <0.01 <0.01 1.00 <0.01 <0.01 1.00 1.00 <0.01

Stimulatory Not significant Stimulatory Not significant Not significant Not significant Not significant Inhibitory Inhibitory Not significant Stimulatory Stimulatory Not significant Not significant Stimulatory

a more gradual increase to its maximum at 72 h. Though DSw BP was the only treatment significantly higher than Cw at 24 h, BP in all treatments (especially surface treatments) increased between 24 and 72 h, with DSw BP continuing to increase by 120 h. DGw was the only treatment that did not exhibit a net stimulatory effect on BP. BP rates in the October incubation were approximately 50 % those in August, although all treatments (except DGw) showed significant stimulation of BP (Table 3). DSw BP increased sharply in the first 24 h and remained high throughout the incubation, while BP increased more gradually in the other treatments. DSw also exhibited increases (~2×) in BA within 24 h (Fig. 4d), but no other significant treatment effects occurred in BA in the October incubation (Table 3). We compared the PP/BP response ratios of treatments at 24 h, relative to controls, to assess whether treatments affected the relative balance of autotrophic and heterotrophic production rates (Fig. 5). In August, groundwater treatment response ratios were not different from the control, but PSw was significantly higher (indicating greater relative stimulation of autotrophs; ANOVA, p <0.05) and DSw was significantly lower (indicating greater relative stimulation of heterotrophs; ANOVA, p <0.05). In October, greater variation existed in response ratios between treatments. The lowest ratios were seen in PGw and DSw, which were significantly lower than those of DGw, PSw and Cw (ANOVA, p <0.05).

August, PP and BP at 24 h did not correlate well with their subsequent performance at 72 h. Interestingly, both primary and bacterial production at 72 h were well predicted by the performance of the reciprocal group at 24 h. Despite a strong treatment response, PP at 24 h did not correlate with any chemical constituent analyzed in either August or October.

Comparing Treatment Responses and Initial Conditions

Fig. 5 The ratio of PP to BP at 24 h after the start of incubations in August (left) and October (right). Letters above columns denote significantly different groups (ANOVA, post-hoc Tukey's test). Though bars and whiskers depict means and 95 % confidence limits of the untransformed data, ANOVAs used to determine significant groups used log transformed data to satisfy normality and variance assumptions

Pearson's product correlations between production rates at either 24- or 72-h time points and initial inorganic nutrients/ organic matter concentrations are shown in Table 4. In

August

a

October

b

Estuaries and Coasts Table 4 Pearson's Correlation Coefficients (n =5) between instantaneous productivity measurements (PP and BP at 24 and 72 h) and initial dissolved nutrient and organic matter concentrations (μM) in incubation bottles in August (top) and October (bottom) Sample time Response 24 h (h) measure PP BP August

24 72

October 24 72

PP BP PP BP PP BP PP BP

72 h PP

0.51 0.46

NH4

NOx

DIN

DON

TDN

SRP

DOP

TDP

DOC

0.38

BP 0.90*

0.01

0.99** 0.83 0.80

0.77 0.79 0.41 0.36 0.42 0.13 −0.23 0.99** 0.95* 0.81 0.93* 0.78 0.90*

0.73

0.69

0.45

0.34

0.17

0.26

0.37

0.91* 0.90** 0.92* 0.57 0.97** 0.98** 0.87

0.94* 0.94** 0.91* 0.38 0.99** 0.99** 0.94*

0.99** 0.99** 0.79 0.23 0.99** 0.98** 0.98**

0.97** 0.98** 0.72 0.08 0.96* 0.93* 0.98**

0.89* 0.92* 0.54 0.59 0.96* 0.97** 0.87

0.92* 0.94* 0.62 0.54 0.98** 0.99** 0.89*

0.94* 0.98** 0.96* 0.98** 0.70 0.74 0.51 −0.39 0.98** 0.67 0.99** 0.67 0.89* 0.78

*p <0.05 **p <0.01 Numbers in bold are significant at the 95% confidence limit.

BP at 24 h, however, was significantly positively correlated with every constituent except for NH4. Analytes that correlated with BP at 24 h had the same relationship with PP at 72 h in both strength and significance. BP at 72 h, on the other hand, remained correlated only with NOx and DIN. In October, 24-h BP was significantly correlated with both PP and BP at 72 h, while PP at 24 h was not a good predictor of either biological rate measurement at 72 h nor did it correlate significantly with any analyte. BP at 24 h correlated with the initial concentrations of all analytes except for NH4 and DOC. Analytes correlating with BP at 24 h again had the same patterns in strength and significance of correlations with initial chemical constituents as PP at 72 h.

Discussion Drainage basin characteristics (urban vs. forested) and hydraulic flowpaths (surface vs. groundwater) can interact to form chemically distinct waters with responses of microplankton communities depend upon these chemical conditions. These interactions not only alter the absolute magnitude of nutrient and organic matter concentrations in basin export waters, they also alter the relative balance of nutrient and organic matter concentrations between the two flowpaths. Surface runoff from the developed basin had substantially higher concentrations of nearly every nutrient and organic compound measured. As a result, it typically yielded the highest production rates and biomass increases when introduced to marine receiving water, with a greater relative stimulation of heterotrophic production relative to autotrophic production. Groundwater from the developed site, on the other hand, was comparatively depleted in nutrient and organic carbon, and this was reflected in a much lower biological response. In fact, this groundwater source had lower nutrient and carbon concentrations,

other than phosphorus, than groundwater from the pristine site. In contrast to developed sites, PGw and surface water were more similar in water chemistry, with respect to both concentrations and relative partitioning, than the developed sources. Since these results are based on only two sampling events in just two drainage basins, we compared the nutrient and organic matter concentrations in our source waters to samples from similar development and flowpath regimes at a regional and national scale. Regionally, Wahl et al. (1997) and Aelion et al. (1997) examined surface and groundwater water quality, respectively, at Oyster Creek (our pristine site) and a developed creek in the nearby city of Murrell's Inlet, SC (a somewhat less urbanized community to the south of our developed site, having a lower percentage of impervious cover). Data from Oyster Creek collected in this study compared well with the previous studies from this site, with the exception of the high NOx concentration measured in our July sampling. It is possible that this may have been the result of the oxidation of groundwaterderived ammonia, as oxic surface water may have mixed with a rising water table prior to sample collection. Comparisons between our developed site and that of Wahl et al. (1997) and Aelion et al. (1997) revealed some substantial differences, however. Our DSw treatment had much higher NH4 and DOC concentrations than the stormwater sampled by Wahl et al. (1997), with our DSw DOC concentrations being 2 times (October) to 5 times (July) greater than the highest stormwater DOC measured by Wahl et al. (1997). In addition, NH4 and NOx concentrations in our DGw were 6–15 times lower than those measured in the developed groundwater sampled by Aelion et al. (1997). Comparisons with national datasets (Fig. 6) suggested that surface runoff from our developed site sampled in this study had nutrient and organic matter concentrations at the upper range of that typical for anthropogenically impacted surface runoff, with concentrations generally above

Estuaries and Coasts

a

b

August

NH 4 NOx DON SRP DOP DOC

NH 4 NO x DON SRP DOP DOC

d Nutrient and Organic Matter Concentration (µM)

c

Surface Runoff

October

Nutrient and Organic Matter Concentration (µM)

Groundwater

Fig. 6 Box and whisker plots of nutrient and organic matter concentrations for groundwater from shallow (<10 m) suboxic (<2 mg/L DO) wells (top; USGS; water.usgs.gov/nawqa/nutrients/ datasets/nutconc2000) and surface runoff (bottom; NSQD; unix.eng.ua.edu/~rpitt/Research/ ms4/mainms4.shtml) derived from developed (urban/ agricultural) landscapes plotted against our source water concentrations used in August (left) and October (right) incubations

NH 4

the 75th and occasionally above the 90th percentiles of the national datasets. Nutrient and organic matter concentrations in DGw and at the pristine site sampled in this study, on the other hand, were generally more similar to previously reported concentrations in the national dataset. One might expect that, given the high concentration of NOx in DSw samples and the solubility of inorganic nitrogen (Pitt et al. 1999), DGw would also have shown elevated nitrogen. DGw was, instead, high in SRP, which has efficient soil adsorption mechanisms. However, sandy soils allow for much deeper penetration of phosphorus than more clayey soils (Pitt et al. 1999), and observations of sandy soil when installing wells at our developed site may pose a possible mechanism for the high phosphorus content in DGw. Denitrification may be a significant process in removing nitrate and nitrite from soils at our study locations, but this process is likely comparable between both developed and pristine sites (Aelion et al. 1997). High nutrient and organic matter content in surface runoff and low concentrations in groundwater appears to be rather common in developed landscapes (Nussbaum 1990), at least for deeper groundwaters (>20 m water depth) and the same may be true of shallow groundwater. This contrasts with observations from saline coastal groundwater, however, which show elevated nutrient

NO x

SRP

NH 4

NO x

SRP

concentrations when collected from areas adjacent to developed land (Àlvarez-Gòngora and Herrera-Silveira 2006; Basterretxea et al. 2010). These elevated concentrations may reflect transformations or mixing dynamics occurring in the subterranean estuary (Moore 1999) due to salt water mixing (Santos et al. 2008), which were presumably not present in the non-saline groundwaters sampled in this study. The strong influence of source water chemical composition on biological response was clearly evidenced by the fact that treatments more similar in their chemical compositions were also more similar in their overall biological responses, regardless of the original source of that water. Overall, surface water sources generally produced greater increases in autotrophic and heterotrophic production, however, than did groundwater sources. In fact, groundwater from the developed watershed seldom stimulated significant biological responses. High nutrient and labile organic matter concentrations in developed stormwater, resulting in significant stimulation of receiving water microbial communities is certainly consistent with the wealth of literature on urban stormwater impacts (e.g., as reviewed by Walsh et al. 2005). The lack of a response with groundwater from the developed watershed, on the other hand, was initially surprising in that it contradicts previous studies of the stimulatory effects of urban groundwater on coastal waters. For example, Garcès et al. (2011)

Estuaries and Coasts

conducted a similar set of groundwater addition experiments in a coastal embayment in the Mediterranean and observed substantial increases in phytoplankton abundances upon as little as 4– 12 % groundwater additions. These additions only slightly increased bacterioplankton abundances, however, leading the authors to conclude that groundwater additions significantly shift autotrophic/heterotrophic balances of the receiving water microbial community. However, the groundwater collected by Garcès et al. (2011) was most likely much more saline and, while similar to previous measurements of groundwater DIN concentrations, was greatly enriched in inorganic nutrients (especially DIN) in comparison to groundwaters used in our study. PP rates were also substantially lower in the October groundwater treatment from the pristine site, despite having high NH4 concentrations (Fig. 4b; Table 2). PGw in October was highly visibly colored relative to other treatments and in comparison to PGw in August (Hutchins, personal observation). We suspect the presence of highly colored dissolved organic matter in this source water likely reduced available photosynthetically active radiation (PAR) in this treatment (Arrigo and Brown 1996), which would explain why this was also the only treatment not to show depletion of NH4 by the end of the incubation. This, potentially coupled with the photolytic breakdown of highly colored organic compounds into more bioavailable substrates for bacteria (Strome and Miller 1978; Smith and Benner 2005), likely decreased the October PP/BP ratio in the PGw treatment, consistent with previous reports of the role of colored dissolved organic inputs on lentic ecosystems (e.g., Pèrez et al. 2003). Both bacteria and phytoplankton consistently exhibited treatment responses in the first 24 h, although the relationships between initial state variables and responses at 24 h were much stronger for BP than for PP (Table 4). This pattern shifted at 72 h, however, when PP was well correlated with most initial water chemistry parameters, while BP became much less correlated with initial conditions. The 72-h time point is also when peak Chl and PP responses were most pronounced in those treatments eliciting significant treatment effects. This suggests that bacterial communities responded quickly to allochthonous nutrients and carbon, while community-level autotrophic responses either required a lag time, as has been seen previously in nutrient addition experiments (e.g., Downing et al.1999), or may have been responding to nutrients regenerated by bacteria (or their primary consumers) from the allochthonous organic material. The drawdown of inorganic nutrient concentrations in all treatments, which was especially rapid in August experiments, is consistent with a high nutrient demand by microbial communities and limitation of phytoplankton production by nitrogen, as has been previously reported for North Inlet waters (Lewitus et al. 1998), which are naturally low in inorganic N and P (Buzzelli et al. 2004). Bacteria also have high nutrient requirements, and have been shown to outcompete phytoplankton for inorganic nutrients, depending on stoichiometry of available

organic matter (Kirchman 1994). Nonetheless, the fact that PP and BP remained elevated in the surface water treatments despite depletion of inorganic nutrient concentrations suggests tight coupling of autotrophy and heterotrophy with PP being driven by regenerated nutrients (Dugdale and Goering 1967). Although not without its caveats (Jahnke and Craven 1995), the ratio of PP/BP provides some insight regarding trophic energy pathways. The PP/BP ratios for both August and October DSw treatments, as well as that for the PGw treatment in October, were significantly lower than control treatments, suggesting the relative stimulation of heterotrophic metabolism over autotrophic metabolism upon addition of these particular source waters. Since source water additions were GF/F filtered (with nominal pore size of 0.7 μm), it is possible that each experimental treatment harbored a slightly different bacterial community composition and that this may have contributed to some of the observed differences in heterotrophic metabolic response (Wear et al. 2013). However, several previous reciprocal transplant studies have shown that metabolic response is much more a function of substrate source and availability than is inoculum source (Kirchman et al. 2004; Xu et al. 2013). In addition, the direct stimulation of heterotrophic metabolism due to allochthonous DOC is well known for freshwater ecosystems (e.g., del Giorgio and Peters 1994; Prairie et al. 2002) and supports the notion that differences among various terrestrial source waters can have significant consequences for the metabolic balance of coastal receiving waters. A shift towards heterotrophic production fueled by allochthous organic sources can alter energy transfer efficiencies within food webs (Jansson et al. 2000) as well as promote lower dissolved oxygen conditions in receiving waters (Verity et al. 2006). Indeed, enhanced heterotrophic metabolism in response to surface runoff from urbanized drainage basins of Myrtle Beach may have some implications for the recent observations of hypoxia and high oxygen consumption rates in coastal waters off this area (Sanger et al. 2012). While our estimates of PP/BP provide a relative index between treatments, interpretations of the specific PP/BP ratio derived from the microcosm measurements must of course be made with caution, as PP rate measurements were conducted at a single light level, rather than taking into account the nonlinear relationships between irradiance and primary production (Jassby and Platt 1976). In summary, our results indicated much greater discrepancies in dissolved nutrient and organic matter concentrations between surface and groundwater flowpaths associated with developed land uses, relative to those of forested land. Although surface runoff generally stimulated greater microbial metabolic responses in coastal receiving waters, as compared to groundwater source inputs, overall biological responses in our microcosms were largely driven by the initial concentration of dissolved constituents irrespective of specific source. If there was any effect at all of differences in the lability of dissolved

Estuaries and Coasts

constituents between treatments, it was largely overshadowed by the differences in absolute concentration. While we must acknowledge the limited scope of our sampling, our results suggest further research regarding the drivers of variability in surface and groundwater concentrations for highly developed coastal watersheds of the Southeastern US is clearly warranted given the potential importance of these differences on biological responses in coastal receiving waters. Acknowledgments Financial support for this work was provided by the National Estuarine Research Reserve Graduate Research Fellowship Program, the Savannah Presbytery, and the Slocum-Lunz Foundation. We thank Ashley Riggs, Amy Willman, Tracy Buck, Jamie Brusa, Damon Tucker and Leigha Peterson for assistance in the lab and in the field; Kevin Godwin and Keshov Jagannathan for help with statistical analyses; Susan Libes, Samantha Joye, and Kimberly Hunter for their perspectives on the experimental design and for providing laboratory access and support. The constructive comments from two anonymous reviewers greatly improved this manuscript.

References Aelion, C.M., J.N. Shaw, and M. Wahl. 1997. Impact of suburbanization on ground water quality and denitrification in coastal aquifer sediments. Journal of Experimental Marine Biology and Ecology 213: 31–51. Agardy, T., and J. Alder. 2005. Coastal systems. In Millennium ecosystem assessment: Current state and trends assessment, ed. R. Hassan, R. Scholes, and N. Ash, 513–548. Washington: Island Press. Allen, J., and K. Lu. 2003. Modeling and prediction of future urban growth in the Charleston region of South Carolina: A GIS-based integrated approach. Conservation Ecology 8: 2. Àlvarez-Gòngora, C., and J.A. Herrera-Silveira. 2006. Variations of phytoplankton community structure related to water quality trends in a tropical karstic coastal zone. Marine Pollution Bulletin 52: 48–60. Arar, E.J., and G.B. Collins. 1997. Method 445.0: In vitro determination of chlorophyll a and pheophytin a in marine and freshwater algae by fluorescence. United States Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory. Arrigo, K.R., and C.W. Brown. 1996. Impact of chromophoric dissolved organic matter on UV inhibition of primary productivity in the sea. Marine Ecology Progress Series 140: 207–216. Azam, F., T. Fenchel, J.G. Field, J.S. Gray, L.A. Meyer-Reil, and F. Thingstad. 1982. The ecological role of water-column microbes in the sea. Marine Ecology Progress Series 10: 257–263. Basterretxea, G., A. Tovar-Sanchez, A.J. Beck, P. Masquè, H.J. Bokuniewicz, R. Coffey, C.M. Duarte, J. Garcia-Orellana, E. Garcia-Solsona, and L. Martinez-Ribes. 2010. Submarine groundwater discharge to the coastal environment of a Mediterranean island (Majorca, Spain): Ecosystem and biogeochemical significance. Ecosystems 13: 629–643. Benner, R., and M. Strom. 1993. A critical evaluation of the analytical blank associated with DOC measurements by high-temperature catalytic oxidation. Marine Chemistry 41: 153–160. Buzzelli, C., Akman, O., Buck, T., Koepfler, E., Morris, J., and Lewitus, A., 2004. Relationships among water-quality parameters from the North Inlet–Winyah Bay National Estuarine Research Reserve, South Carolina. Journal of Coastal Research: 59–74 Crossett, K.M., Culliton, T.J, Wiley, P.C. and T.R. Goodspeed. 2004. Population trends along the coastal United States: 1980–2008. NOAA Coastal Trends Series. Available from: http://oceanservice. noaa.gov/programs/mb/pdfs/coastal_pop_trends_complete.pdf (accessed August 1, 2012)

Daley, R., and S. Brown. 1973. Experimental characterization of lacustrine chlorophyll diagenesis: I. Physiological and environmental effects. Archives of Hydrobiology 72: 277–304. del Giorgio, P.A., and R.H. Peters. 1994. Patterns in planktonic P:R ratios in lakes: Influence of lake trophy and dissolved organic carbon. Limnology and Oceanography 772–787. DiDonato, G.T., J.R. Stewart, D.M. Sanger, B.J. Robinson, B.C. Thompson, A.F. Holland, and R.F. Van Dolah. 2009. Effects of changing land use on the microbial water quality of tidal creeks. Marine Pollution Bulletin 58: 97–106. Downing, J.A., C.W. Osenberg, and O. Sarnelle. 1999. Meta-analysis of marine nutrient-enrichment experiments: Variation in the magnitude of nutrient limitation. Ecology 80: 1157–1167. Dugdale, R.C., and J.J. Goering. 1967. Uptake of new and regenerated forms of nitrogen in primary productivity. Limnology and Oceanography 12: 196–206. Eaton, A.D., and Franson, M.A.H.. 2005. Standard methods for the examination of water & wastewater. Amer Public Health Assn. Epps, T.H., D.R. Hitchcock, A.D. Jayakaran, D.R. Loflin, T.M. Williams, and D.M. Amatya. 2012. Characterization of storm flow dynamics of headwater streams in the South Carolina lower coastal plain. Journal of the American Water Resources Association 49: 76–89. Garcès, E., G. Basterretxea, and A. Tovar-Sànchez. 2011. Changes in microbial communities in response to submarine groundwater input. Marine Ecology Progress Series 438: 47–58. Garside, C. 1982. A chemiluminescent technique for the determination of nanomolar concentrations of nitrate and nitrite in seawater. Marine Chemistry 11: 159–167. Grimm, N.B., S.H. Faeth, N.E. Golubiewski, C.L. Redman, J.G. Wu, X.M. Bai, and J.M. Briggs. 2008. Global change and the ecology of cities. Science 319: 756–760. Harder, S.V., D.M. Amatya, T.J. Callahan, C.C. Trettin, and J. Hakkila. 2007. Hydrology and water budget for a forested Atlantic coastal plain watershed, South Carolina. Journal of the American Water Resources Association 43: 563–575. Hitchcock, J.N., S.M. Mitrovic, T. Kobayashi, and D.P. Westhorpe. 2010. Responses of estuarine bacterioplankton, phytoplankton and zooplankton to dissolved organic carbon (DOC) and inorganic nutrient additions. Estuaries and Coasts 33: 78–91. Holland, A.F., D.M. Sanger, C.P. Gawle, S.B. Lerberg, M.S. Santiago, G.H.M. Riekerk, L.E. Zimmerman, and G.I. Scott. 2004. Linkages between tidal creek ecosystems and the landscape and demographic attributes of their watersheds. Journal of Experimental Marine Biology and Ecology 298: 151–178. Jahnke, R.A., and D.B. Craven. 1995. Quantifying the role of heterotrophic bacteria in the carbon cycle: A need for respiration rate measurements. Limnology and Oceanography 40: 436–441. Jansson, M., A.-K. Bergström, P. Blomqvist, and S. Drakare. 2000. Allochthonous organic carbon and phytoplankton/bacterioplankton production relationships in lakes. Ecology 81: 3250–3255. Jassby, A.D., and T. Platt. 1976. Mathematical formulation of the relationship between photosynthesis and light for phytoplankton. Limnology and Oceanography 21: 540–547. Kirchman, D.L. 1993 Leucine incorporation as a measure of biomass production by heterotrophic bacteria. In: Kemp, P.F., B.F. Sherr, E.B. Sherr and J.J. Cole (eds) Handbook of methods in aquatic microbial ecology. Lewis Publishers, p 513–517 Kirchman, D. 1994. The uptake of inorganic nutrients by heterotrophic bacteria. Microbial Ecology 28: 255–271. Kirchman, D., J. Sigda, R. Kapuscinski, and R. Mitchell. 1982. Statistical analysis of the direct count method for enumerating bacteria. Applied and Environmental Microbiology 44: 376–382. Kirchman, D.L., A.I. Dittel, S.E.G. Findlay, and D. Fischer. 2004. Changes in bacterial activity and community structure in response

Estuaries and Coasts to dissolved organic matter in the Hudson River, New York. Aquatic Microbial Ecology 35: 243–257. Kjerfve, B. 1986. Circulation and salt flux in a well mixed estuary. Physics of Shallow Estuaries and Bays 16: 22–29. Knap, A., A. Michaels, A. Close, H.W. Ducklow, and H. Dickson. 1996. Protocols for the Joint Global Ocean Flux Study (JGOFS) core measurements. Report no. 19, Reprint from the IOC Manuals and Guides no. 29. Bergen: UNESCO. Koroleff, F. 1983. Simultaneous oxidation of nitrogen and phosphorus compounds by persulfate. Methods of Seawater Analysis 2: 168–169. Lapointe, B.E., J.D. O'Connell, and G.S. Garrett. 1999. Nutrient couplings between on-site sewage disposal systems, groundwaters, and nearshore surface waters of the Florida Keys. Biodegradation 10: 289–307. Lawrenz, E., E.M. Smith, and T.L. Richardson. 2013. Spectral irradiance, phytoplankton community composition and primary productivity in a salt marsh estuary, North Inlet, South Carolina, USA. Estuaries and Coasts 36: 347–364. Lewitus, A.J., E.T. Koepfler, and J.T. Morris. 1998. Seasonal variation in the regulation of phytoplankton by nitrogen and grazing in a saltmarsh estuary. Limnology and Oceanography 43: 636–646. Mallin, M.A., V.L. Johnson, and S.H. Ensign. 2009. Comparative impacts of stormwater runoff on water quality of an urban, a suburban, and a rural stream. Environmental Monitoring and Assessment 159: 475–491. Mallin, M.A., K.E. Williams, E.C. Esham, and R.P. Lowe. 2000. Effect of human development on bacteriological water quality in coastal watersheds. Ecological Applications 10: 1047–1056. Moore, W.S. 1999. The subterranean estuary: A reaction zone of ground water and sea water. Marine Chemistry 65: 111–125. Nussbaum, J.C. 1990. Urban storm water runoff and ground-water quality. Office of the Administrator, United States Environmental Protection Agency Paerl, H.W., J. Dyble, P.H. Moisander, R.T. Noble, M.F. Piehler, J.L. Pinckney, T.F. Steppe, L. Twomey, and L.M. Valdes. 2003. Microbial indicators of aquatic ecosystem change: Current applications to eutrophication studies. FEMS Microbiology Ecology 46: 233–246. Paul, M.J., and J.L. Meyer. 2001. Streams in the urban landscape. Annual Review of Ecology and Systematics 32: 333–365. Pèrez, A.P., M.M. Diaz, M.A. Ferraro, G.C. Cusminsky, and H.E. Zagarese. 2003. Replicated mesocosm study on the role of natural ultraviolet radiation in high CDOM, shallow lakes. Photochemical and Photobiological Sciences 2: 118–123. Pitt, R., S. Clark, and R. Field. 1999. Groundwater contamination potential from stormwater infiltration practices. Urban Water 1: 217–236. Prairie, Y.T., D.F. Bird, and J.J. Cole. 2002. The summer metabolic balance in the epilimnion of southeastern Quebec lakes. Limnology and Oceanography 47: 316–321. Sanger, D.M., A.F. Holland, and G.I. Scott. 1999. Tidal creek and salt marsh sediments in South Carolina coastal estuaries: II. Distribution of organic contaminants. Archives of Environmental Contamination and Toxicology 37: 458–471. Sanger, D.M., E.M. Smith, G. Voulgaris, E.T. Koepfler, S.M. Libes, G.H.M. Riekerk, D.C. Bergquist, D.I. Greenfield, P.A. Wren, C.A. McCoy, R.F. Viso, R.N. Peterson, and J.D. Whitaker. 2012. Constrained enrichment contributes to hypoxia formation in Long Bay, South Carolina (USA), an open water urbanized coastline. Marine Ecology Progress Series 461: 15–30. Santos, I.R., W.C. Burnett, J. Chanton, B. Mwashote, I.G.N.A. Suryaputra, and T. Dittmar. 2008. Nutrient biogeochemistry in a Gulf of Mexico subterranean estuary and groundwater-derived fluxes to the coastal ocean. Limnology and Oceanography 53: 705–718. Santos, I.R., R.N. Peterson, B.D. Eyre, and W.C. Burnett. 2010. Significant lateral inputs of fresh groundwater into a stratified

tropical estuary: Evidence from radon and radium isotopes. Marine Chemistry 121: 37–48. Scheiner, S.M., and J. Gurevitch. 2001. Design and analysis of ecological experiments. USA: Oxford University Press. Schindler, D., R. Schmidt, and R. Reid. 1972. Acidification and bubbling as an alternative to filtration in determining phytoplankton production by the 14C method. Journal of the Fisheries Board of Canada 29: 1627–1631. Siegel, A., B. Cotti-Rausch, D.I. Greenfield, and J.L. Pinckney. 2011. Nutrient controls of planktonic cyanobacteria biomass in coastal stormwater detention ponds. Marine Ecology Progress Series 434: 15–27. Slomp, C.P., and P. Van Cappellen. 2004. Nutrient inputs to the coastal ocean through submarine groundwater discharge: Controls and potential impact. Journal of Hydrology 295: 64–86. Smith, D.C., and F. Azam. 1992. A simple, economical method for measuring bacterial protein synthesis rates in seawater using 3Hleucine. Marine Microbial Food Webs 6: 107–114. Smith, E.M., and R. Benner. 2005. Photochemical transformations of riverine dissolved organic matter: Effects on estuarine bacterial metabolism and nutrient demand. Aquatic Microbial Ecology 40: 37–50. Solorzano, L. 1969. Determination of ammonia in natural waters by the phenolhypochlorite method. Limnology and Oceanography 14: 799– 801. Solorzano, L., and J.H. Sharp. 1980. Determination of total dissolved phosphorus and particulate phosphorus in natural waters. Limnology and Oceanography 25: 754–758. Strome, D., and M. Miller. 1978. Photolytic changes in dissolved humic substances. In Proceedings of 20th Congress , Internationale Vereinigung fur Theoretische und Angewandte Limnologie, 1248– 1254. Denmark. Swarzenski, P.W., C. Reich, K.D. Kroeger, and M. Baskaran. 2007. Ra and Rn isotopes as natural tracers of submarine groundwater discharge in Tampa Bay, Florida. Marine Chemistry 104: 69–84. Tufford, D.L., C.L. Samarghitan, H.N. McKellar, D.E. Porter, and J.R. Hussey. 2003. Impacts of urbanization on nutrient concentrations in small southeastern coastal streams. Journal of the American Water Resources Association 39: 301–312. Van Dolah, R.F., G.H.M. Riekerk, D.C. Bergquist, J. Felber, D.E. Chestnut, and A.F. Holland. 2008. Estuarine habitat quality reflects urbanization at large spatial scales in South Carolina's coastal zone. Science of the Total Environment 390: 142–154. Verity, P.G., M. Alber, and S.B. Bricker. 2006. Development of hypoxia in well-mixed subtropical estuaries in the southeastern USA. Estuaries and Coasts 29: 665–673. Vitousek, P.M., H.A. Mooney, J. Lubchenco, and J.M. Melillo. 1997. Human domination of Earth’s ecosystems. Science 277: 494–499. Wahl, M., H. McKellar, and T. Williams. 1997. Patterns of nutrient loading in forested and urbanized coastal streams. Journal of Experimental Marine Biology and Ecology 213: 111–131. Walsh, C.J., A.H. Roy, J.W. Feminella, P.D. Cottingham, P.M. Groffman, and R.P. Morgan. 2005. The urban stream syndrome: Current knowledge and the search for a cure. Journal of the North American Benthological Society 24: 706–723. Wear, E.K., E.T Koepfler and E.M. Smith. 2013. Spatiotemporal variability in dissolved organic matter composition is more strongly related to bacterioplankton community composition than to metabolic capability in a blackwater estuarine system. Estuaries and Coasts. In press. Woodward, R.T., and Y.S. Wui. 2001. The economic value of wetland services: A meta-analysis. Ecological Economics 37: 257–270. Xu, J., H. Jing, L. Kong, M. Sun, P.J. Harrison, and H. Liu. 2013. Effect of seawater–sewage cross-transplants on bacterial metabolism and diversity. Microbial Ecology 66: 60–72.