Estuaries and Coasts (2012) 35:325–334 DOI 10.1007/s12237-011-9433-3
Inorganic and Organic Nitrogen Uptake by Phytoplankton and Bacteria in Hong Kong Waters Xiangcheng Yuan & Patricia M. Glibert & Jie Xu & Hao Liu & Mianrun Chen & Hongbin Liu & Kedong Yin & Paul J. Harrison
Received: 11 January 2011 / Revised: 24 June 2011 / Accepted: 29 July 2011 / Published online: 2 September 2011 # Coastal and Estuarine Research Federation 2011
Abstract Measurements of uptake rates of inorganic (NO3− and NH4+) and organic (urea, glycine, and glutamic acid) N, and indirect estimates of total N uptake by bacteria, were made in four contrasting environments in sub-tropical Hong Kong waters in summer of 2008. In addition, the effects of several days of rain on N uptake rates were studied in eastern waters. Although ambient NO3− was the dominant form of N in Hong Kong waters, the dominant N form taken up by phytoplankton was usually NH4+ and organic N, including urea and amino acids, rather than NO3−. Hence, because of the low NO3− uptake, there was a long turnover time for NO3− (100 days), and most of the NO3− was apparently transported offshore into deeper shelf waters. In eastern waters where NH4+ was undetectable, NO3− uptake rates were positively correlated with phytoplankton cell size. In contrast, potential rates of glutamic acid uptake were negatively correlated with phytoplankton size. N uptake rates X. Yuan (*) : K. Yin State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China e-mail:
[email protected] X. Yuan : J. Xu : H. Liu : M. Chen : H. Liu : P. J. Harrison Division of Environment, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China P. M. Glibert Horn Point Laboratory, University of Maryland Center for Environmental Science, PO Box 775, Cambridge, MD 21613, USA K. Yin School of Marine Sciences, Zhong Shan (Sun Yat-Sen) University, Guangzhou, China
in the smaller size fraction (0.7–2.8 μm) were less affected by the rain event, and smaller phytoplankton appeared to outcompete larger cells after several days of rain. The surface (PN)-specific N uptake rates in the >8-μm fraction decreased from 0.02 to 0.0001 h−1, while the smaller fraction only exhibited a one- to threefold decrease after the rainfall. In contrast, bacterial production and N uptake were not affected by the rain event, and bacteria N uptake accounted for 10– 60% of the total N uptake by phytoplankton. Keywords N uptake . Size-fractionated phytoplankton . Bacteria . Coastal waters . Pearl River estuary
Introduction As the population in southern China continues to increase, inorganic and organic nutrient loading, especially nitrogen (N), has increased due mainly to domestic and industrial sewage discharge into the Pearl River estuary (PRE) and adjacent waters (Yin and Harrison 2007; Xu et al. 2008). Hong Kong is located on the eastern side of the Pearl River estuary and provides an interesting site to address many of the issues related to N pollution. Victoria Harbour in Hong Kong receives more than 2 million tons of sewage effluent daily (Choi et al. 2009). Previous studies suggested that anthropogenic inputs from the Pearl River discharge and sewage effluent have led to increases in inorganic nutrients and organic matter as well as phytoplankton biomass in the PRE and adjacent waters (Yin and Harrison 2007; Yuan et al. 2010). The mean annual rainfall is about 2,413 mm year−1, based on the 1986–2001 average (Ho et al. 2008), and it contributes to the nutrient loading while wind events result in vertical mixing in these shallow waters.
326
To date, there have been several studies that have addressed the regulation of primary and bacteria production by phosphorus (P) as it has been considered to be the potential limiting nutrient for these Pearl River influenced waters in the summer (Yin and Harrison 2007; Xu et al. 2008; Yuan et al. 2011). However, there has been little emphasis on the uptake of N by either phytoplankton or bacteria. Moreover, compared to the numerous studies of such rates conducted in temperate coastal and estuarine waters (Glibert et al. 1982; Glibert and Garside 1992; Wafar et al. 2004; L’Helguen et al. 2008), sub-tropical and tropical waters have been comparatively little studied. In this study, the uptake rates of inorganic (NH4+, NO3−) and organic N (urea and glycine and glutamic acid) by phytoplankton and indirect estimates of N utilization by bacteria in the Hong Kong waters were determined. A rainfall event during the study also afforded the opportunity to contrast responses before and during the event. Our objective was to assess the relative use of different forms of N by phytoplankton and bacteria in different regions of Hong Kong waters differentially impacted by Pearl River and other N loads.
Estuaries and Coasts (2012) 35:325–334
Nutrients, Particulate Nitrogen, and Chlorophyll a Nutrient samples were filtered through precombusted GF/F glass fiber filters and immediately frozen until analyzed. Inorganic nutrient concentrations were determined colorimetrically (Knap et al. 1994) with a SKALAR autoanalyzer. NO3− was analyzed by the Cu-Cd column reduction method and NH4+ with the indophenol blue color formation, respectively. Urea was measured with diacetyl monoxime thiosemicarbazide (Price and Harrison 1987). No measurements were made of concentrations of individual amino acids. All resulting rate calculations of amino acid uptake rates should be considered to be relative only and are likely overestimates. Samples for particulate nitrogen (PN) were filtered through precombusted (460°C, 2 h) GF/F filters and determined using a CHN analyzer. The PN of different plankton size fractions (as outlined below) was not measured on July 3, only total PN was measured. Chl a was filtered onto Whatman GF/F glass fiber filters, extracted with 90% acetone, and analyzed using a fluorometer (Knap et al. 1994). Inorganic and Organic N Uptake Rates
Materials and Methods Sampling Sampling was conducted in western, southern, and eastern Hong Kong waters in July 2008 (Fig. 1). Western waters are influenced by the freshwater discharge from the Pearl River estuary, while southern waters are influenced by a mixture of freshwater, sewage effluent, and coastal/ shelf water. Victoria Harbour is near the local sewage discharge site, and eastern waters are considered to be a reference station since they receive little anthropogenic input from the Pearl River estuary or the local sewage discharge site. The western, southern, and Victoria Harbour sites were sampled at depths corresponding to five light levels (1%, 10%, 30%, 50%, and 100% of surface light), while most sampling in the eastern site was near surface. At the eastern site, an additional experiment was conducted to study N uptake by different plankton size fractions before a rain event on July 3 and following several days of rain on July 7. Photosynthetically available radiation (PAR) in the water column was measured using a Li-Cor underwater spherical quantum sensor (LI 193SA, USA), while the solar radiation in the air was measured using a Li-Cor pyranometer (LI-200SZ, USA). The vertical profiles of salinity and temperature were measured with a YSI® 6600 sensor.
Uptake rates of inorganic and organic N were measured using 15N-labeled NH4+, NO3−, urea, and amino acids (glycine and glutamic acid). 15N substrates were added to whole water samples at concentrations that were ∼10% of ambient values. An initial sample from each incubation sample was also filtered onto a precombusted GF/F filter and the filtrate retained for nutrient analysis. Although five depths were sampled, the surface samples were incubated at 100% and 50% light; the midwater column samples were incubated at 30% and 10% light, and the near bottom samples were incubated at 1% light. Duplicate incubations were conducted on selected samples to determine the coefficient of variation (CV) of N uptake, and the CV of N uptake was mostly within 8%. The temperature of the ondeck incubator was maintained within ±2°C of the in situ water temperature by running seawater. Sample bottles were incubated for ∼1–2 h and then filtered onto precombusted GF/F filters directly or as described below. The 15N samples were analyzed by mass spectrometry on a Sercon mass spectrometer. PN-specific and absolute uptake rates were calculated using the equations of Glibert and Capone (1993). Different Size Fractions Two experiments were conducted on N uptake by plankton size fractions before a rain event on July 3 and following a period of rain on July 7. The total amount of rain on July 3,
327
Fig. 1 Map and sampling stations in western waters (WW), southern waters (SW), Victoria Harbour (VH), and eastern waters (EW)
23
Estuaries and Coasts (2012) 35:325–334
China
22.5
22.50
Shenzhen
Latitude ( oN)
22
Hong Kong
Sewage discharge site 113
113.5
114
114.5
115
WW
EW 22.25
VH
SW 113.75
114.00
114.25 Longitude ( E)
114.50
o
4, 5, 6, and 7 was 0, 0, 11, 54, and 39 mm day−1, respectively, while wind speeds were 11, 8, 15, 25, and 24 km h−1, respectively (http://www.hko.gov.hk). Samples from eastern waters were size-fractionated by filtering through a filter tower fitted with Nitex screening (pore size 8 μm), Whatman GF/D glass fiber filter (nominal pore size 2.7 μm), and Whatman GF/F glass fiber filter (nominal pore size 0.7 μm) on July 3. The Nitex screening was then gently washed onto a GF/F filter. All glass fiber filters were precombusted (450oC, 2 h) before use. This effectively divided the phytoplankton into >8 μm (netplankton), 2.7– 8 μm (nanoplankton), and 0.7–2.7 μm (picoplankton and some bacteria). Bacterial Abundance and Production and Primary Production Bacterial abundance was determined by flow cytometry. Samples (1.8 ml) were preserved with 1% paraformaldehyde + 0.05% glutaraldehyde (final concentration) and stored at −80°C. The samples were later thawed, stained with 0.01% SYBR Green I (Molecular Probes) in the dark at room temperature for 20 min before analysis, and run through a Becton Dickinson FACS Calibur cytometer. Data were obtained in log mode until ∼100,000 events were acquired. Bacteria were detected by their signature in a plot of side scatter versus FL1 (green fluorescence) as outlined in Gasol and del Giorgio (2000). The resulting cytograms were analyzed using Cytowin 4.3 software. The precision for estimates of bacterial abundance was generally within 2% (CV). Bacterial production (BP) was determined as described by Simon and Azam (1989). 3H-leucine (final concentration 30 nM; specific activity 55.9 Ci mmol−1) was added to 2 ml
subsamples (triplicate) with one control fixed by 5% trichloroacetatic acid (TCA). The subsamples were incubated for leucine incorporation for 1 h and the linearity of the incorporation of leucine was tested in a separate time series experiment (data not shown). The incubation was terminated by adding TCA (5% final concentration). After centrifugation and aspiration of the supernatant, pellets were rinsed and centrifuged twice with 1 ml of 5% TCA, and then scintillation cocktail was added to the vial. The incorporated 3H was determined using a Beckman® 1414 CA/LL liquid scintillation counter. BP was calculated using an empirical conversion factor (CF) of ∼3 kg C mol per leucine (Yuan et al. 2010). For primary production, seawater from the 100%, 50%, and 1% light depths was incubated in 50 ml bottles for 2– 4 h under natural solar radiation after an addition of 10– 20 μCi NaH14CO3 in dim light, both with different neutral density screening to reduce surface light to 50% and 1%. The incubation was terminated by filtering through a 25 mm Whatman GF/F filter and the filters were kept frozen (−80°C) until they were analyzed within 1 month following the Joint Global Ocean Flux Study (JGOFS) protocols (Knap et al. 1994). The filters were placed into scintillation vials containing 0.2 ml of 0.5 N HCl for 12 h in order to remove inorganic carbon. After the addition of 10 ml of scintillation cocktail (Hi-Safe) to each vial, samples were counted in a Beckman® 1414 CA/LL liquid scintillation counter and 14C uptake rates were calculated according to the JGOFS protocols (Knap et al. 1994). Calculations N uptake rates were calculated according to Glibert and Capone (1993). The amino acid concentrations were not
328
Estuaries and Coasts (2012) 35:325–334
waters (∼0–15%) (Fig. 3b), since NH4+ concentrations were low (from detection limits to 1 μM) in Victoria Harbour and eastern waters. The concentrations of PO43− and SiO42− exhibited clear spatial variability, with the highest (>1.2 μM PO43− and>60 μM SiO42−) in western waters, intermediate (0.2–1.0 μM PO43− and 20–52 μM SiO42−) in Victoria Harbour and southern waters, and the lowest (0.05, insignificant at this level
NH4 (μM)
Slope
r2
n
p
Slope
r2
n
p
0.01 0.009 *
0.3 0.4 *
25 25 25
0.06 0.05 *
0.002 0.058 *
0.6 0.9 *
25 25 25
0.01 0.001 *
Estuaries and Coasts (2012) 35:325–334 Turnover time (days) 0
50 100 150 200 0
50 100 150 200 0
50 100 150 200 0
50 100 150 200 0 10 20 30 40 50 0 10 20 30 40 50
50 30
NO3 NH4
10
Urea
SW (Jul 4)
Discussion Our results showed that there were relatively high concentrations of ambient inorganic N and uptake rates by phytoplankton and bacteria in Hong Kong waters. In addition,
VH (Jul 2)
VH (Jul 4)
Turnover times of Nitrogen In Victoria Harbour, the high NO3− concentrations and low uptake rates resulted in a long turnover time of NO3− in western waters. The water residence times were ∼1.5– 2.5 days in summer (Kuang and Lee 2004), which was much shorter than the turnover times of the NO3−, suggesting that much of the NO3− would be transported onto shelf
A)
July 7 (after the rainfall) Surface Bottom 3
B)
-1
0.03
0.03
0.02
2
0.01
1
0.00
0
0.02 0.01 0.0
100
C)
100 80
60
60
40
40
20
20
0
0
>8 2.8 0.7 -8 -2. 8
80
NH4 Urea Amino acids
Chl a PN
D)
E)
5
0.03
4
0.02
3
0.01
2
0.00
1
> 2. 8 8 0 . -8 72. 8
NO3
0.04
N uptake rates -1 (μM N h )
>8 2.8 0.7 8 -2. 8
Relative N uptake (%)
PN-specific -1 uptake rates (h )
EW (Jul 7)
the different forms of N exhibited different concentrations and uptake rates by phytoplankton and bacteria, which were influenced by episodic events (e.g., rainfall). The relative use of different forms of N by phytoplankton affected the turnover time and the fate of N.
July 3 (before the rainfall) Surface Bottom 0.04
EW (Jul 3)
Chl a (μg l )
SW (Jul 2)
Phytoplankton size (μm)
> 2. 8 8 0 . -8 72. 8
WW (Jul 2)
PN (μΜ)
1
(glutamic acid + urea) accounted for a larger amount (∼10% to 60%) of the total N uptake in eastern waters when there was little NH4+ available (Fig. 7c). The smaller-sized plankton (e.g., smaller phytoplankton and bacteria associated with particles) apparently utilized more organic N compared to the larger cells on July 3, but this was not the case on July 7 after the rain event (Fig. 7c, f).
Fig. 7 PN-specific uptake rates and chl a concentrations on July 3 (a) and July 7 (b); relative N uptake of different forms of N (NO3−, NH4+, urea, and glutamic acid) on July 3 (c) and July 7 (d); and particulate nitrogen (PN) and absolute N uptake rates on July 7 for three plankton size fractions (>8, 2.8– 8, and 0.7–2.8 μm) at the surface and bottom (100% and 1% light depths) in eastern waters (EW) (e)
50 100 150 200 0
100
% Light
Fig. 6 Turnover times of three nitrogen sources (NO3−, NH4+, and urea) at five different light levels in western waters (WW) on July 2, southern waters (SW), and Victoria Harbour (VH) on July 2 and 4 and eastern waters (EW) on July 3 and 7. Note the different scale on the y-axis for EW on July 3 and 7
331
332
waters. Long turnover times of NO3− (19–2,160 days) have also been reported in other turbid and tidal estuaries (Shaw et al. 1998; Middelburg and Nieuwenhuize 2000). NH4+ was nearly depleted in Victoria Harbour during our sampling, despite the high sewage discharge in this area. The high uptake rates and low ambient NH4+ concentrations (near the detection limit) resulted in short turnover times of NH4+ uptake in Victoria Harbour and eastern waters. Turnover times of NH4+ in Hong Kong were in the range of those reported for other tidal coastal waters (0.1– 27 days; Shaw et al. 1998; Middelburg and Nieuwenhuize 2000). In contrast to summer, previous studies showed that the discharge of sewage effluent resulted in high NH4+ (7 to 20 μM) in Victoria Harbour and its vicinity in winter when biological uptake is low due to low light and temperature (Yin and Harrison 2007). Dominant Forms of N Taken up by Phytoplankton Ambient NO3− concentrations were the highest among all forms of N in Hong Kong waters (Xu et al. 2008), but NO3− uptake rates were often the lowest. In contrast, lower ambient NH4+ concentrations supported high rates of NH4+ uptake. In southern waters, NH4+ was the dominant form of N taken up on July 2 and July 4 when NH4+ was depleted, while organic N was the main N form taken up at the surface (Figs. 5a and 6b). Similarly, in Victoria Harbour and eastern waters where NH4+ was near the detection limit, organic N was often the dominant form of N taken up (Fig. 5a). Only when NH4+ was low and decreased to detection limits did NO3− uptake became dominant, which suggested that ambient NH4+ concentrations influenced the sequence of N forms taken up. Previous studies also showed that the ambient concentrations at which NH4+ can reduce or partially inhibit NO3− uptake vary greatly between phytoplankton species (Glibert et al. 1982; Lomas and Glibert 1999; Collos et al. 2004) and with N status before NH4+ was added (Dortch et al. 1991). However, the interaction between inorganic and organic N has been little investigated, although a laboratory experiment suggests that the repression of NO3− uptake rates by NH4+ is mediated by glutamine (Page et al. 1999). Amino acids have traditionally been ignored as a source of N for autotrophs, despite evidence that they directly use dissolved free amino acids (e.g., leucine, glutamine, and glycine) to varying degrees (Wheeler et al. 1974; Wheeler 1983; Flynn and Butler 1986; Berman and Bronk 2003). Our results showed that amino acids may have potentially represented as much as ∼60% of phytoplankton N uptake in eastern waters (Fig. 5c), which was much higher than the highest values (∼20%) reported in the mid-Atlantic Bight off southern New Jersey and Chesapeake Bay (Bradley et al. 2010a, b), but these values must be interpreted cautiously due to the potential for overestimation.
Estuaries and Coasts (2012) 35:325–334
N Uptake by Size-Fractionated Phytoplankton and Effects of a Rain Event Primary production was 2 to 40 μM C h−1 (or ∼0.5 to 2 g C m−3 day−1) at the surface in this study, which was within the range of ∼0.5 to 2.7 g C m−3 day−1 in the wet season during 2005–2006 (Ho et al. 2010). The chl a biomass and primary productivity in surface waters are mostly dominated by the 5–20-μm size fraction, mainly fastgrowing chain-forming diatoms in the wet season in Hong Kong waters (Ho et al. 2010; Xu et al. 2008). Larger phytoplankton (>20 μm) only accounted for a small portion of the phytoplankton biomass (16:1) have also been reported in other studies (Paerl et al. 1999; Herut et al. 1999). Hence, P likely became even more limiting for phytoplankton after the rain event. During the rain event, high wind speeds (24 km h−1) resulted in a deeper mixed layer depth (∼5 m) than before (∼2 m). The highest PAR was ∼900 μmol m−2 s−1 at the surface (∼1 m) before the rain event and decreased to 1% of surface light at ∼5 m. Although PAR was not measured during the rain event, daily solar radiation decreased from ∼25 to 4.5 MJ m−2 during the rain on July 7, suggesting that PAR might also decrease and result in light limitation, especially a few meters below the surface. Smaller phytoplankton have an advantage over larger cells under light limitation, owing to their higher surface-tovolume ratio (Riegman and Noordeloos 1998), which may be one of the explanations for higher chl a in the small size fraction (
