Article pubs.acs.org/est
Acceleration of Denitrification in Turbid Rivers Due to Denitrification Occurring on Suspended Sediment in Oxic Waters Ting Liu,† Xinghui Xia,†,* Shaoda Liu,‡ Xinli Mou,† and Yiwen Qiu† †
School of Environment, Beijing Normal University/State Key Joint Laboratory of Environmental Simulation and Pollution Control, Beijing, 100875, China ‡ Department of Geography, National University of Singapore, 1 Arts Link, Kent Ridge, 117570, Singapore S Supporting Information *
ABSTRACT: High suspended sediment (SPS) concentration exists in many rivers of the world. In the present study, the effects of SPS concentration on denitrification were investigated in airtight chambers with sediment samples collected from the Yellow River which is the largest turbid river in the world. Results from the nitrogen stable (15N) isotopic tracer experiments showed that denitrification could occur on SPS in oxic waters and the denitrification rate increased with SPS concentration; this was probably caused by the presence of low-oxygen microsites in SPS. For the water systems with both bed-sediment and SPS, the denitrification kinetics fit well to Logistic model, and the denitrification rate constant increased linearly with SPS concentration (p < 0.01). The denitrification caused by the presence of SPS accounted for 22%, 38%, 53%, and 67% of the total denitrification in systems with 2.5, 8, 15, and 20 g L−1 SPS, respectively. The activity of denitrifying bacteria in SPS was approximately twice that in bed-sediment, and the denitrifying bacteria population showed an increasing trend with SPS concentration in both SPS and bed-sediment, leading to the increase of denitrification rate with SPS concentration. Furthermore, the denitrification in bed-sediment was accelerated by increased diffusion of nitrate from overlying water to bed-sediment under agitation conditions, which accompanied with the presence of SPS. When with 8 g L−1 SPS, approximately 66% of the increased denitrification compared to that without SPS was attributed to denitrification on SPS and 34% to agitation conditions. This is the first report of the occurrence of denitrification on SPS in oxic waters. The results suggest that SPS plays an important role in denitrification in turbid rivers; its effect on nitrogen cycle should be considered in future study.
■
INTRODUCTION Nitrogen is one of the most important nutrients for organisms, serving as a major ingredient for synthesis of animal protein and growth of plants. However, excessive nitrogen discharged into environment will lead to water eutrophication, water hypoxia, reduction in species diversity, and disruption of ecosystems and water habitat. Denitrification, which converts nitrate (NO 3−-N) to dinitrogen (N2) or nitrous oxide (N2O), is the major pathway for permanent removal of nitrogen out of water bodies.1−4 Bedsediments are commonly considered important for denitrifying bacteria to eliminate nitrogen from river systems, because the anoxic environment and organic matter in bed-sediments can provide the conditions required for denitrification.5−7 However, some studies have shown that the denitrification efficiency in bed-sediment is usually regulated by nitrate diffusion from water into bed-sediment and limited by the surface area of bedsediment. 8−10 Furthermore, Billen et al. 11 found that denitrification in oxygen minimum zones of highly heterotrophic water-column, where nitrate is directly available to bacteria, is much more efficient in removing NO3−-N than that © XXXX American Chemical Society
in bed-sediment because of less transport limitations for nitrate within water than that from water into bed-sediment. It is reported that denitrifying bacteria are mainly associated with particles in water columns.12 Many researchers suggest that anaerobic microsites might exist in suspended particles of oxygenated waters based on the theories of multispecies biofilm and transport limitations.12−15 In addition, because of the existence of anaerobic microsites in some particles, simultaneous nitrification and denitrification is found in fluidized bed bioreactor (FBBR) systems, which use very small particles as the support for biofilm growth and have been widely used in wastewater treatment.16−20 Therefore, we hypothesize in the present study that denitrification would occur on suspended sediment (SPS) in natural waters, and SPS may be more advantageous for denitrification than bed-sediments due to the easy access of Received: November 5, 2012 Revised: March 26, 2013 Accepted: April 2, 2013
A
dx.doi.org/10.1021/es304504m | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
denitrifying bacteria in SPS to nitrate. Additionally, our previous research showed that nitrification rate increased with SPS concentration in river water due to the increase of bacteria population with SPS concentration.21,22 Therefore, we also hypothesize that the denitrification rate would increase with increasing SPS concentration. High SPS concentration exists in many rivers of the world. For instance, the Yellow River in China has an average SPS concentration of 22 g L−1;23 the SPS concentration of the Yangtze River ranges from 0.1 to 26 g L−1 in the river mouth.24 The Mississippi River, the fourth longest river in the world, has a SPS concentration of 5.1 g L−1 at St. Louis, MO,25 and the SPS concentration of the Delaware River in Kansas ranges from 0.1 to 11.7 g L−1.26 Despite the universal existence of high SPS concentration in river water, there is no research report about the effect of SPS on denitrification in rivers with high SPS concentration. The Yellow River, with a drainage area of 7.52 × 105 km2, is the second longest river in China. Nitrogen pollution is one of the most critical problems in the Yellow River,22,27,28 which is the main reason for the eutrophication in the Bohai bay.29 Here we took the Yellow River, the largest turbid river in the world, as a case to study the effect of SPS concentration on denitrification. The objectives of the present study were (1) to investigate whether denitrification could occur on SPS in oxic waters through isotopic tracer experiment conducted in water systems containing only SPS (SPS-water systems (SPS-WS)); (2) to compare the denitrification rate under different SPS concentrations; (3) to explore the mechanisms regarding the effect of SPS concentration on denitrification; (4) to determine the contribution of denitrification occurring on SPS to total denitrification in water systems containing both SPS and bedsediment (bed sediment-SPS-water systems (BS-SPS-WS)) and to assess the importance of denitrification on SPS in turbid rivers.
Figure 1. Laboratory experiment chamber constructed from a PMMA column.
Denitrification Simulation Experiment. Effect of SPS Concentration on Denitrification in SPS-WS. To investigate whether denitrification could occur on SPS, a set of experiment containing water and SPS but no bed-sediment was designed. A total of 800 mL water containing 5 mg L−1 15NO3−-N as K15NO3 (99.0 atm% 15N, Shanghai Research Institute of Chemical Industry, China) was added into a series of chambers respectively, obtaining a 20 cm deep water column. Then the chambers were added with 2, 6.4, 12, and 16 g of homogenized bed-sediment, respectively. The chambers were incubated at 25 °C, and the agitation rates were set at 200, 400, 550, and 700 r min−1 to make all the added sediment suspend, with SPS concentrations of 2.5, 8, 15, and 20 g L−1, respectively. In addition, one set of chambers containing only 800 mL water extraction solution of sediment with SPS of 0 g L−1 was designed to investigate denitrification in oxygen-saturated water phase, and agitation rate was set at 50 r min−1. At predetermined intervals, gas samples were collected by a gastight syringe to detect the δ15N−N2 produced during each incubation period; then the water columns were aerated for 30 min to purge the old 15N2 of previous incubation period from the chamber and keep oxygen saturation during experiment. At the end of incubation, water and sediment were separated by standing, and denitrifying bacteria density in the water and SPS samples were measured using most-probable-number (MPN)PCR method. Denitrification rate was calculated according to the production of 15N2. Each experimental set was incubated in triplicate, and the control experiments were carried out with sterilized water and sediment samples, and microbial activities were inhibited by adding 0.5% mercuric chloride. Effect of SPS Concentration on Denitrification in BS-SPSWS. A known amount of homogenized bed-sediment was added to each chamber to obtain a 10 cm deep bed-sediment column, then it was covered with 800 mL artificial overlying water containing 5 mg L−1 NO3−-N as KNO3. The SPS concentrations were set at 0, 2.5, 8, 15, and 20 g L−1 by adjusting the agitation rate at 0, 200, 400, 550, and 700 r min−1, respectively, mimicking the real rivers with different SPS concentrations, and the variation of actual SPS concentration in each system was less than 5% of the preset one (details of SPS
■
MATERIAL AND METHODS Sample Collection. Water and sediment samples were collected at Huayuankou Hydrological Station (113°41′07.7″E, 34°54′16.8″N) near Zhengzhou in the middle reach of the Yellow River, where monthly average SPS concentration varied between 3 and 24 g L−1.23 To avoid contaminating samples by any surface pollutants, water and SPS samples were collected at 0.2 m below the surface in the midstream using a TC-Y sampler (TECH Instrument in Shenyang, China). Bed-sediment samples were collected from the top 10 cm of bed-sediment using a sediment grab sampler. All water and bed-sediment samples were kept under 4 °C in a cooler and transported to laboratory for experimental analysis. Lab experiments began within 48 h after sampling. Design of Incubation Chamber. A special chamber was designed for laboratory experiment. As shown in Figure 1, the chamber is made up of a polymethyl methacrylate (PMMA) column (7 cm inner diameter and 40 cm height), whose end is sealed by a rubber stopper. Three glass tubes pass through the rubber stopper for aeration and collection of water and gas samples. A mechanical agitator, fixed on the stopper, is used to stir water and produce different concentrations of SPS (details shown in Supporting Information (SI)). Prior to experiment, each chamber was checked for air tightness by pumping air out with a vacuum pump, and the pressure in each chamber was decreased to 0.9 atm and could keep at this pressure for 24 h. B
dx.doi.org/10.1021/es304504m | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
samples were reported in standard δ notation (‰) relative to atmospheric N2. Precision of the δ15N measurement was ±0.2‰. The calculation method about the production of 15 N2−N from 15NO3−-N was described in the SI. Size characteristic of SPS was measured with a laser particle sizer (Microtrac, S3500). Moisture content of sediment was determined by calculating weight difference between wet sediment and that dried at 100 °C for 4h. Total organic carbon (TOC) in sediment was determined with the hightemperature K2Cr2O7 method,30 and that in water was determined with a TOC analyzer (Shimadzu TOC-500). Temperature, pH, and redox potential of water were measured with a pH meter (Mettler Toledo, SG23). Dissolved oxygen content of water was measured with an oxygen meter (Mettler Toledo, SG9-FK2). Biological Analysis. DNA was extracted from sediment before and after incubation experiments to analyze denitrifying bacteria abundance. All extractions were conducted in duplicate with an UltraClean Soil DNA Isolation Kit (Mo-bio Laboratories, Inc., Carlsbad, CA) according to the manufacturer’s protocols. The extracted DNA was then stored at −20 °C until analysis by MPN-PCR.31 A 10-fold serial dilution of the extracted DNA was then prepared, in triplicate, for the amplification process. PCR amplification of nirS fragment, which encode for nitrite reductase of denitrifying bacteria, was conducted using the 832F (5′-TCACACCCCGAGCCGCGCGT-3′) and 1606R (5′-AGKCGTTGAACTTKCCGGTCGG-3′) primer set.32 All PCR amplifications were carried out with 40 μL of reaction mixture (20 mM Tris-HCl, 50 mM KCl) containing 1.5 mM MgCl2, 0.2 mM each deoxyribonucleotide triphosphate, 0.1 μM of the primer, and 1.25U of Taq polymerase (Takara); after an initial denaturation step at 94 °C for 5 min, 30 cycles of 94 °C denaturation (30 s), 58 °C annealing (45 s), 72 °C elongation (1.5 min) were followed by a final elongation step of 10 min at 72 °C. A total of 5 μL of the amplicons was visualized on agarose gels (1% TAE) stained with Goldview nucleic acid staining (SBS Genetech, China), and the presence of a DNA band corresponding to the expected size was treated as a positive result. Enumeration of denitrifying bacteria was calculated according to the number of positive amplifications per dilution and with the help of Cochran tables.33 Quality Assurance and Quality Control. The quality assurance and quality control (QA/QC) included (a) The ranges of recoveries for NH4+-N, NO2−-N, NO3−-N and organic-N in the sediment and water samples were 96−104%, 96−102%, 95−102%, and 95−105%, respectively; the relative standard deviations were less than 4.4%, 2.8%, 5.4%, 5.0%, respectively. (b) The variations of NH4+-N, NO2−-N, NO3−-N, and organic-N were less than 5% in the control experiments with sterilized water and sediment samples; the 15N2 emission in the control experiments was less than 7% of that in the systems with bacteria. (c) The denitrification rate was calculated according to the 15N2 emission or the NO3−-N variations in systems. As organic-N and NH4+-N might transform into NO3−-N during the incubation, the calculated denitrification rate based on NO3−-N might be less than the real denitrification rate in the system, but this error was less than 10% based on the variations of organic-N and NH4+-N in the system.
determination is shown in SI). These chambers were incubated at 25 °C. At predetermined intervals, water and sediment (including both bed-sediment and SPS) samples were collected using a columnar sampler which is a hollow narrow cylinder that can dip into the bottom of bed-sediment and suck up samples, and the samples were put into tubes to separate water and sediment by centrifuge; then water and sediment samples were examined for the levels of nitrogen species. After each sampling, the water phase of each chamber was aerated for 30 min to keep oxygen saturation during experiment. At the end of incubation experiment, the overlying water including SPS was decanted, and stood for 12 h to allow SPS to deposit. The denitrifying bacteria density in water phase, SPS and bedsediment samples were detected by MPN-PCR method. Denitrification rate was calculated based on the variation of nitrate concentration in the systems including water, SPS and bed-sediment phases, with nitrate concentration expressed as mg NO3−-N per liter water. Each experimental set was incubated in triplicate with a set of control. The control experiments were carried out with sterilized water and bedsediment samples, and microbial activities were inhibited by adding 0.5% mercuric chloride. Another experimental set was conducted to investigate NO3−-N gradient in pore water of bed-sediment (details shown in SI). In addition, we took the SPS concentration of 8 g L−1, which appeared in the temperate season (March−June and October− November) in the Yellow River, as an example to investigate the contribution of denitrification caused by SPS and turbulence to the total denitrification in BS-SPS-WS. Three kinds of systems were designed: (A) system containing only 8 g L−1 SPS as mentioned in the above section (Section 2.3.1); (B) system containing bed-sediment covered by static water (bed sediment-water systems (BS-WS)); (C) system containing both bed-sediment and 8 g L−1 SPS. The experimental procedures including incubation, sampling and analysis were the same as that mentioned in the above section except that a 10 cm deep bed-sediment was constructed in systems (B) and (C), and the water phase of system (B) was agitated gently with agitation rate of 50 r min−1 for 10 min to make 15N2 release from the water to gas phase before gas sampling. Denitrification rate was calculated according to the production of 15N2. All the above experimental conditions were summarized and shown in SI Table S1. Sample Analysis. Chemical Analysis. Water sample collected from each chamber was filtered through a 0.45 μm filter, and analyzed for NH4+-N, NO2−-N, and NO3−-N concentration immediately. Sediment was extracted with 2 M KCl for 30 min on a shaking table and centrifuged at 3000g for 15 min, and then the supernatant was collected and analyzed for NH4+-N, NO2−-N, and NO3−-N concentration in sediment. Dissolved nitrogen (NH4+-N, NO2−-N, and NO3−-N) concentrations were measured colorimetrically with an Autoanalyser-3 (Bran & Luebbe, France). Organic nitrogen (Organic-N) in the water and sediment was calculated by subtracting the ammonia value from the total Kjeldahl nitrogen, and the latter was obtained with an Autoanalyser-3 by measuring the acid digest samples which were prepared by Kjeldahl digestion of water or sediment samples in a Block Digestor (BD50s, Bran & Luebbe, France). All chemical reagents used in analysis were of analytical grade. The 15N2 in gas samples were analyzed using an isotopic ratio mass spectrometer (Thermo Delta V Advantage), in line with an automated gas PreCon unit. Nitrogen isotope ratios of C
dx.doi.org/10.1021/es304504m | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
■
Article
RESULTS AND DISCUSSION The properties of sediment sample are shown in Table 1, according to which 81% of the sediment were particles ranging
The denitrifying bacteria population in SPS was at least 2 orders of magnitude higher than that in water phase for all systems with different SPS concentrations at the end of incubation (SI Figure S1), indicating that denitrifying bacteria tended to grow on SPS. The denitrifying bacteria population in SPS increased with SPS concentration (Figure 2B), and the average denitrification rate was significantly correlated with them (r = 0.990, p < 0.01, SI Figure S2 ). For the system with 15 g L−1 SPS, the denitrification rate only increased a little compared to that with 8 g L−1 SPS; this was probably caused by the little increase of denitrifying bacteria population in the system. Although some researchers isolated aerobic denitrifiers from wastewater treatment systems, pond waters and paddy sediments,36−41 most of the bacteria conduct denitrification under limited-oxygen conditions (
