1 Introduction

Increases of nutrient concentrations in aquatic ecosystem are the main reason for eutrophication (Smith 2003). Eutrophication leads to the impairment of water quality and deterioration of aquatic ecosystem, resulting in biodiversity decline and economic losses (Dodson et al. 2000; Smith 2003). As a result, many research focus on nutrients reduction to mitigate the eutrophication problem (Carpenter 2008; Jarvie et al. 2013). Although many measures are carried out to cut off the external loading, nutrient concentrations might still be high in water due to internal release from the contaminated sediment (Jeppesen et al. 2005; Søndergaard et al. 2007). Therefore, internal nutrient loadings need to be properly controlled for a successful restoration attempt.

Many restoration techniques have been developed and used in the past decades to control internal loadings. Sediment dredging is a simple way to remove internal nutrients but the outcomes were found controversial (Jing et al. 2013; Liu et al. 2015; Zhang et al. 2010). In addition, high cost and problems associated with the treatment and disposal of the dredged sediment prevent the wide application of sediment dredging (Murphy et al. 1999). In situ treatment, which can considerably reduce the cost and cause less disturbance to the sediment, becomes a more attractive solution.

In situ treatment with chemical modifiers can be an effective way to control internal nutrient loadings. Chemical modifiers such as aluminum, iron salts, nitrate, and Phoslock® have been studied in previous research (Hansen et al. 2003; Liu et al. 2009; Meis et al. 2013). Aluminum and iron salts were able to inhibit the release of phosphorus (P) from sediment by combining with soluble reactive phosphorus (SRP) to form insoluble precipitants (Liu et al. 2009). However, aluminum can be toxic to aquatic organisms (Pessot et al. 2014), and the effect of iron salt is sensitive to redox and pH conditions (Burley et al. 2001; Immers et al. 2013).

Nitrate application was firstly tested by Ripl (1976), which achieved a drastic reduction of the nutrient levels and an improvement of the redox condition. Research had shown that nitrate addition could not only eliminate black-odorous and suppress internal phosphorus loading in the contaminated sediment but also successfully stimulate indigenous microorganisms for in situ bioremediation (Hemond and Lin 2010; Xu et al. 2014; Yamada et al. 2012). Phoslock®, a new lanthanum-modified bentonite, was developed in the 1990s by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) of Australia (Robb et al. 2003), and had succeeded in reducing phosphorus in both laboratory and field trials (Meis et al. 2012; Meis et al. 2013; Lürling and van Oosterhout 2013). Phoslock® had also been used to remove phosphorus from wastewater and to control phosphate mobilization in soils (Geurts et al. 2011; Haghseresht et al. 2009).

However, the phosphorus immobilization mechanisms by these two modifiers and their effects on nitrogen (N) cycling in sediment has not been well characterized. Therefore, this work aims to (1) examine the changes of phosphorus in water and sediment over time following the treatment with calcium nitrate and Phoslock®, (2) to evaluate their potential impacts on physicochemical properties of the sediment and on N cycling, and (3) to assess and compare the modes of action of these two modifiers.

2 Materials and Methods

2.1 Materials and Chemicals

Sediment used in this work was collected from Donghu Lake (30° 33′ N, 114° 23′ E), the largest urban lake in China located in the city of Wuhan with an area of 33 km2 and an average depth of 2.5 m. Over 100 small-scale enterprises; 11 hospitals; and 200,000 residential homes are located around the lake (Wang et al. 2002). Due to the discharge of municipal wastewater, the phosphorus dynamics showed an upward trend with a peak period in the mid-1980s in the lake accompanied by an outbreak of cyanobacteria blooms (Xie and Xie 2002). The input of wastewater has been considerably reduced in recent years as a result of the city sewer system upgrade; however, the internal P load will continue to threat the water quality of the lake. Sediment sample was collected from the lake (30° 32′ 54.91″ N, 114° 21′ 15.61″ E) with a stainless steel grab sampler, homogenized, and sieved to <0.18 mm to remove the coarse particles (to avoid damage to the microelectrode). Water content, total organic carbon (TOC), total Kjeldahl nitrogen (TKN), and total phosphorus (TP) of the sediment was 68.34 %, 1.67 %, and 3.89 mg N/g and 2.32 mg P/g, respectively. The concentrations of NH4 +–N, NO3 −–N, and NO2 −–N were 0.23, 0.21, and 0.07 mg N/g, respectively.

Calcium nitrate (Ca(NO3)2·4H2O) is analytical reagent grade and was provided by Sinopharm Chemical Reagent Co., Ltd. Phoslock® (lanthanum-modified bentonite containing 5 % lanthanum) was supplied by the Phoslock® Water Solutions Ltd. All other chemicals used in the experiment are analytical reagent grade and were purchased from Sinopharm Chemical Reagent Co., Ltd.

2.2 Microcosm Experiment

Microcosms were setup in 1-L beakers with a height of 19 cm and a diameter of 10 cm. Each beaker was filled with 800-g sediment and 600-mL water, and the heights of sediment and the overlaying water were about 10 and 8 cm, respectively. In treatment 1, calcium nitrate was added to the surface of the sediment at a dose of 61 g N/m2. In treatment 2, 4 g of Phoslock® (equal to approximately 510 g/m2) was added by sprinkling the solid evenly onto the surface of the sediment. In control, no modifier was added. The application rate was selected by reference to the literature. Previous research by Foy (1986) showed that the addition of 61 g N/m2 of nitrate to the sediment delayed and reduced phosphorus release, causing complete suppression. Intact core experiment performed by Meis et al. (2013) indicated that 510 g Phoslock®/m2 is required to control P release under anaerobic condition.

Microcosms were maintained at 24 °C in dark. Three beakers were retreated from each treatment and control on days 1, 15, 30, and 66. Overlaying water was siphoned into a sampling bottle, and the sediment was sliced into 2-cm sections to a depth of 8 cm using a column sampler. The remaining sediment in the beaker was homogenized for the measurement of water content. The sliced sediment samples from each layer from each treatment and control were mixed to form composited samples.

On days 1, 5, 30, and 66, a microelectrode (Unisense, Denmark) was used to analyze the dissolved oxygen (DO) at the sediment-water interface. After about an hour of polarization and calibration, the electrode was emplaced onto a micromanipulator (MM 33, Unisense). The electrode was 1 cm above the surface of the sediment. Motor controller (MC-232, Unisense) drove the electrode down 500 μm every 5 s to a depth of 7 mm below the sediment surface.

2.3 Chemical Analysis

Water pH was measured by a Hach miniLab H135 pH meter. Water samples were filtered through 0.45-μm cellulose acetate membranes. Total nitrogen (TN) and TP of the overlaying water were measured using the alkaline persulfate oxidation and persulfate oxidation method, respectively. NH4 +–N was determined by Nessler’s colorimetric, NO2 −–N and soluble reactive phosphorus (SRP) were determined by N-(1-Naphthalene)-diaminoethane and ammonium molybdate methods, and NO3 −–N was determined by the UV spectrophotometry method (MEPC 2002).

Water content of each sample was measured from the weight loss by drying at 105 °C to a constant weight. Total Kjeldahl nitrogen (TKN) of the sediment was determined by the automatic Kjeldahl nitrogen analyzer (S5, Behr, Germany). TKN includes ammonia and organic N which could be transformed to ammonia via sulfuric acid digestion. NH4 +–N and NO2 −–N were measured by spectrophotometric method following potassium chloride extraction. NO3 −–N was determined by phenol disulfonic acid spectrophotometric method. TN was calculated as the sum of TKN, NO2 −–N, and NO3 −–N. TP was measured as described by Ruban et al. (1999).

The sequential extraction scheme of Psenner et al. (1984) with slightly modifications was used for P fractionation. P was separated into loosely sorbed P (NH4Cl-P), reductant-soluble P (BD-P), metal oxide bound P (NaOH-SRP), organic bound P (NaOH-NRP), calcium bound P (HCl-P), and residual-P. NaOH-NRP was calculated as NaOH-TP minus NaOH-SRP, and residual-P was calculated as TP substrate all the extractable P species.

2.4 Data Analysis

Results of the overlaying water samples were presented as the means and standard deviations of three replicates. Results of the sediment samples were reported on dry weight basis. Differences between treatments and control for NO3 −–N, NO2 −–N, NH4 +–N, TP, and SRP in overlaying water were tested using one-way ANOVA by a LSD post hoc test using PASW Statistics 18 (SPSS Inc.). A confidence level of 95 % was used.

3 Results and Discussion

3.1 The Influences of the Modifiers on Physical and Chemical Conditions

3.1.1 Effects on pH and Water Content

Chemical treatment (e.g., Al or Fe salts) may cause a decline in pH, which may cause a re-release of the phosphorus. In our study, the pH in the overlaying water in calcium nitrate treatment was slightly lower than the control (p < 0.05) (Table 1), and this might be due to the release of hydrogen ion during the oxidation of reducible agents (e.g., Fe salts) by nitrate (Ottley et al. 1997). Possible reactions involved in this process are as follows:

$$ 8{\mathrm{Fe}}^{2+}+{{\mathrm{NO}}_3}^{-}+13{\mathrm{H}}_2\mathrm{O}\to 8\mathrm{FeOOH}+{{\mathrm{NH}}_4}^{+}+14{\mathrm{H}}^{+} $$
(1)
$$ 8{\mathrm{Fe}}^{2+}+{{2\mathrm{NO}}_3}^{-}+11{\mathrm{H}}_2\mathrm{O}\to 8\mathrm{FeOOH}+{\mathrm{N}}_2\mathrm{O}+14{\mathrm{H}}^{+} $$
(2)
$$ 10{\mathrm{Fe}}^{2+}+{{2\mathrm{NO}}_3}^{-}+14{\mathrm{H}}_2\mathrm{O}\to 10\mathrm{FeOOH}+{\mathrm{N}}_2+18{\mathrm{H}}^{+} $$
(3)
$$ 2{\mathrm{Fe}}^{2+}+{{\mathrm{NO}}_3}^{-}+3{\mathrm{H}}_2\mathrm{O}\to 2\mathrm{FeOOH}+{{\mathrm{NO}}_2}^{-}+4{\mathrm{H}}^{+} $$
(4)
$$ 12{\mathrm{Fe}}^{2+}+{{\mathrm{NO}}_3}^{-}+13{\mathrm{H}}_2\mathrm{O}\to 4{\mathrm{Fe}}_3{\mathrm{O}}_4+{{\mathrm{NH}}_4}^{+}+22{\mathrm{H}}^{+} $$
(5)
$$ 12{\mathrm{Fe}}^{2+}+{{2\mathrm{NO}}_3}^{-}+11{\mathrm{H}}_2\mathrm{O}\to 4{\mathrm{Fe}}_3{\mathrm{O}}_4+{\mathrm{N}}_2\mathrm{O}+22{\mathrm{H}}^{+} $$
(6)
$$ 15{\mathrm{Fe}}^{2+}+{{2\mathrm{NO}}_3}^{-}+13{\mathrm{H}}_2\mathrm{O}\to 5{\mathrm{Fe}}_3{\mathrm{O}}_4+{\mathrm{N}}_2+28{\mathrm{H}}^{+} $$
(7)
$$ 3{\mathrm{Fe}}^{2+}+{{\mathrm{NO}}_3}^{-}+3{\mathrm{H}}_2\mathrm{O}\to {\mathrm{Fe}}_3{\mathrm{O}}_4+{{\mathrm{NO}}_2}^{-}+4{\mathrm{H}}^{+} $$
(8)
Table 1 Changes of pH value in the overlaying water and water content in the sediment (mean ± standard deviation, n = 3) during the experiment
Full size table

The Phoslock® treatment has a slightly higher pH (p < 0.05) in the first few days but showed no difference (p > 0.05) at the end compared with the control (Table 1). Ross et al. (2008) found that Phoslock® was most effective at pH values between 5 and 7, and the adsorption capacity decreased above pH 9. Therefore, the slight change of pH here wouldn’t affect the adsorption capacity of Phoslock®. A significant increase (p < 0.05) in water content of the nitrate-treated sediment was also observed after day 15 (Table 1). This was likely attributed to the formation of N2 and N2O in denitrification processes, which made the sediment loose (Zhang et al. 2009). The denitrification processes can be represented by the following two equations:

$$ {{8\mathrm{NO}}_3}^{-}+5{\mathrm{S}}^{2-}+8{\mathrm{H}}^{+}\to 4{\mathrm{N}}_2+{{5\mathrm{SO}}_4}^{2-}+4{\mathrm{H}}_2\mathrm{O} $$
(9)
$$ {{8\mathrm{NO}}_3}^{-}+3{\mathrm{S}}^{2-}+8{\mathrm{H}}^{+}\to 4{\mathrm{N}}_2\mathrm{O}+{{3\mathrm{SO}}_4}^{2-}+4{\mathrm{H}}_2\mathrm{O} $$
(10)

3.1.2 DO at the Sediment-Water Interface

The sediment-water interface was divided into aerobic layer (from the sediment surface and above), transition layer, and anaerobic layer (from the place where DO drops to 0 and below). As presented in Fig. 1, the addition of nitrate had a great influence on DO content. At the beginning, the application of nitrate caused a decline in DO in the aerobic and transition layers on day 1 and 5. However, similar trend in two treatments and control was observed on day 30. On day 66, DO in calcium nitrate treatment was even higher than that in control in the aerobic layer, and the boundary between the transition lay and anaerobic layer increased from 2 mm below the sediment surface on day 1 to 6 mm.

Fig. 1
figure 1

Variation of dissolved oxygen concentration overtime in the profile of the sediment-water interface

Full size image

The addition of calcium nitrate could promote denitrification process in the sediment (Eqs. (9) and (10)), resulting in a release of Fe2+ due to the oxidation of metal sulfide to sulfate (Shao et al. 2009). The mobilized Fe (II) diffused into the sediment-water interface, and when they contact with oxygen or nitrate, Fe (II) was oxidized (Schauser et al. 2006). Nitrate can easily penetrate into the sediment, it was possible that most of Fe (II) was oxidized to Fe (III) by nitrate before they entered the water from sediment. So, there might be other responsible reasons for this result. Dissolved organic carbon (DOC) might enter the overlaying water from sediment due to the loose of the sediment. Yamada et al. (2012) found that nitrate addition could increase the DOC in both interstitial water and overlaying water a few days after treatment. Therefore, it was possible that the DO depletion was due to the oxidation of DOC. Over time, DO in the overlaying water increased gradually as the decrease of DO consumption. Nitrate addition increased the redox potential of the sediment. Therefore, less DO might be consumed during the diffusion to deeper layers. As a result, the depth of the transition layer increased.

DO in aerobic layer was slightly higher in Phoslock® treatment than in control on days 5 and 66. Phoslock® transformed SRP into highly recalcitrant P forms and controlled P release from sediment (Meis et al. 2013). This process can reduce the bioavailable P and as a consequence, might reduce microbial respiration (Ramírez et al. 2003). Therefore, the increase of DO concentration of Phoslock® treatment might relate to the decreased DO consumption by microbes. These results indicated that both nitrate and Phoslock® can improve the redox condition in sediment. However, the effect of calcium nitrate was more pronounced.

3.2 Effects of the Modifiers on Phosphorus

3.2.1 Effects on Phosphorus in the Overlaying Water

Variation of SRP and TP in the overlaying water during the experiment is presented in Fig. 2. SRP in Phoslock® treatment decreased on day 1 and maintained at low concentration during the experiment. SRP in calcium nitrate treatment decreased on day 1 but increased gradually thereafter, and at the end of the experiment, SRP was comparable to its initial concentration, which was different from the results from other researchers (Ripl 1976; Yamada et al. 2012). Release of SRP and TP from porewater might have occurred in calcium nitrate treatment as a result of water content increase due to gas production from denitrification in the sediment. In control, SRP increased first and then dropped to even lower than the initial concentration. TP showed a similar trend as SRP in Phoslock® treatment and control. Whereas TP decreased in days 1 and 15 and then increased on day 30 and then returned back. Overall, Phoslock® showed a better effect in removing P from water and controlling P release from sediment.

Fig. 2
figure 2

Variation of soluble reactive phosphorus (SRP) (a) and total phosphorus (TP) (b) concentrations overtime in the overlaying water

Full size image

Phoslock® could reduce SRP by formation of highly insoluble lanthanum–phosphate complex (Eq. (11)), and then striped it from water column to sediment (Haghseresht et al. 2009; Reitzel et al. 2013).

$$ {\mathrm{La}}^{3+}+{{\mathrm{PO}}_4}^{3-}\to {\mathrm{La}\mathrm{PO}}_4\downarrow $$
(11)

The addition of calcium nitrate could transform Fe (II) to Fe (III) (Eqs. (1)–(4)), which can bind SRP from water (McAuliffe et al. 1998; Yamada et al. 2012). In addition, calcium ion could also bind with P. This might be why SRP and TP decreased in the first few days. However, subsequent increase in SRP and TP might be related to the mineralization of organic materials. In addition, P can be released into the overlaying water during the release of gases produced during the denitrification processes in the sediment.

3.2.2 Effects on Phosphorus in the Sediment

Changes of P species in the sediment profile over time are presented in Fig. 3. Shift in P species was observed following calcium nitrate and Phoslock® application. Phoslock® application mainly affected the top layer of the sediment (0~2 cm), while nitrate treatment can have a deeper influence, mainly the first three layers (0~6 cm). Slight change in P species in control was also observed during the experiment, mainly in the top layer of the sediment.

Fig. 3
figure 3

Changes of phosphorus species over time in the sediment profile: NH4Cl-P (a), BD-P (b), NaOH-SRP (c), NaOH-NRP (d), HCl-P (e), and residual-P (f)

Full size image

NH4Cl-P represents loosely sorbed P, which includes P in porewater, CaCO3-associated P, or P leaching from decaying cells of bacterial biomass in deposited phytodetrital aggregates (Kaiserli et al. 2002). NH4Cl-P in the control increased slight over time in the top layer. NH4Cl-P in the Phoslock® treatment showed little variation over time and was lower than the control in the top layer, while NH4Cl-P increased over time and was higher than the control in almost all layers (Fig. 3a). The most drastic change in nitrate treatment occurred in the top layer, in which NH4Cl-P increased from 4.08 mg P/kg on day 1 to 39.9 mg P/kg on day 66. Increase of NH4Cl-P in nitrate treatment was likely attributed to the increase of P releasing from decaying cells of bacterial biomass in deposited phytodetrital aggregates as nitrate addition might increase the microbial activity in the sediment (Xu et al. 2014).

BD-P represents the redox-sensitive P forms, mainly bound to Fe and Mn compounds (Kozerski and Kleeberg 1998; Schauser et al. 2006). BD-P showed a similar trend as NH4Cl-P. In the top layer, BD-P increased in two treatments and control but increase was more drastic in calcium nitrate treatment, in which BD-P increased from 0.29 mg P/g on day 1 to 0.92 mg P/g day 66 (Fig. 3b). The change of BD-P might be attributed to oxidation of Fe (II) to Fe (III) (Eqs. (1)–(4)) as a result of improved redox condition as observed in Fig. 1. Previous research showed that addition of nitrate promoted autotrophic denitrification, leading to the oxidation of iron sulfide and release of Fe (II) into water (Yamada et al. 2012). Fe (II) was then oxidized and the formation of Fe oxide-hydroxide provided the adsorption sites for P, leading to the increase of BD-P.

NaOH-SRP represents P bound to metal oxides, mainly of Al and Fe, which is exchangeable again with OH− and inorganic P compounds soluble in alkaline condition (Kozerski and Kleeberg 1998). NaOH-NRP represents P in microorganisms including poly-P, organic P in detritus, P bound in humic compounds (Hupfer et al. 1995). Nitrate application decreased NaOH-SRP, and slightly increased NaOH-NRP on days 15 and 30, while Phoslock® application slightly decreased NaOH-SRP in the top layer (Fig. 3c, d). This result suggests that calcium nitrate addition can transform NaOH-SRP into other P species, especially BD-P, which increased most. Anion exchange between nitrate and phosphate may be responsible for the NaOH-SRP decrease. The increase of NaOH-NRP on days 15 and 30 might be due to the increase of microbial biomass stimulated by N addition. With the decrease of N in the sediment as a result of denitrification, NaOH-NRP decreased as well. In Phoslock® treatment, part of NaOH-SRP may be transformed into HCl-P, which is more stable.

HCl-P represents P bound to carbonates, apatite-P, and traces of hydrolyzed organic P (Hupfer et al. 1995). HCl-P increased from 0.38 mg P/g on day 1 to 0.57 mg P/g on day 66 due to the addition of Phoslock®, and this increase was mainly observed in the top layer of the sediment (Fig. 3e). Whereas, calcium nitrate showed minor influence on HCl-P. This result suggests that P can be released from lanthanum–phosphate complex at acid condition. But at typical environmental conditions, release of HCl-P is unlikely to happen. Therefore, HCl-P can be considered as non-bioavailable.

Residual-P is refractory P, which is difficult to release, and can be treated as an inert P-fraction (Psenner and Pucsko 1988). Residual-P increased in nitrate treatment and control (Fig. 3f). It was likely that NaOH-SRP was partially transformed into residual-P pool in calcium nitrate treatment, which might be related to the microbial processes.

NH4Cl-P, BD-P, and NaOH-SRP were considered as mobile P fractions and can be potentially released under certain environment conditions (Hupfer et al. 1995; Meis et al. 2012). Whereas, NaOH-NRP, HCl-P, and residual-P were generally non-bioavailable and have a low release potential (Hupfer et al. 1995; Gonsiorczyk et al. 1998; Kaiserli et al. 2002). Our results showed that the application of Phoslock® transformed mobile P into un-bioavailable P. Calcium nitrate application transformed NaOH-SRP into NH4Cl-P, BD-P, and residual-P. Therefore, the effects of calcium nitrate in P control are more sensitive to the variation of environmental conditions.

3.3 Effects of the Modifiers on N Cycling

3.3.1 Effects on the N Cycling in the Overlaying Water

The addition of Phoslock® showed minor effect on N species. Only NH4 +–N showed a significant increase on day 1 (p < 0.05). Following the addition of calcium nitrate, the overlaying water showed a sharp rise in NO3 −–N concentration (Fig. 4a). The highest NO3 −–N concentration appeared on day 1 and was 718.20 mg/L. However, it decreased over time and dropped to 42.40 mg/L by the end of the experiment, which reduced by 94 % compared to the highest level. Nitrate addition significantly increased (p < 0.05) the concentration of NH4 +–N and NO2 −–N in the overlaying water. The highest concentration of NO2 −–N was observed on day 15 with a value of 2.60 mg/L (Fig. 4b), and the highest NH4 +–N concentration was observed on day 1 with a value of 0.93 mg/L (Fig. 4c). However, NO2 −–N and NH4 +–N decreased over time after peaking with no significant different from the control at the end of the experiment.

Fig. 4
figure 4

Variation of NO3 −–N (a), NO2 −–N (b), and NH4 +–N (c) concentrations overtime in the overlaying water

Full size image

The decline of nitrate and increase of NO2 −–N was mainly attributed to denitrification, and the increase of NH4 +–N might be related to the dissimilatory nitrate reduction to ammonia process (DNRA). The added nitrate acted as electron acceptor, leading to the enhancement of two biological processes including sulfide-oxidizing and organics-oxidizing (Chen et al. 2013). In sulfide-oxidizing process, nitrate might cause the development of nitrate-reducing, sulfide-oxidizing bacterium (NR-SOB) thus enhanced denitrification (García de Lomas et al. 2006). The increase of NH4 +–N and NO2 −–N in nitrate treatment was temporary with the consumption of DO at the beginning of the experiment as described in Section 3.1.2, NH4 +–N can be oxidized to nitrate by facultative heterotrophic microorganisms. NO2 −–N increased in the first few days as an intermediate product of DNRA and denitrification processes. With the consumption of nitrate, the concentration of NO2 −–N decreased as well.

Accumulate of NH4 +–N in Phoslock® treatment on day 1 might be related to the suppression of the nitrification process caused by the capping of bentonite on the sediment surface (Gibbs et al. 2011). However, this influence seems to be transient, with the improvement of redox condition, NH4 +–N was soon oxidized.

3.3.2 Effects on the N Cycling in the Sediment

The addition of Phoslock® showed minor effects on NH4 +–N, NO2 −–N, and NO3 −–N in sediments. NO3 −–N concentrations were low during the whole experiment period, which ranged from 0 to 1.60 mg N/kg. Calcium nitrate treatment showed influence on NO2 −–N and NO3 −–N. NO3 −–N content declined gradually with depth in the sediment. The highest NO3 −–N concentration appeared in the top layer on day 1 with a value of 2023 mg N/kg. By the end of the experiment, it declined by 98 % to 35.59 mg N/kg (Fig. 5a).

Fig. 5
figure 5

Changes of NO3 −–N (a), NO2 −–N (b), and NH4 +–N (c) concentrations overtime in the sediment profile

Full size image

As an intermediate product in NO3 −–N reduction process, NO2 −–N in the sediment also increased before day 15 but decreased thereafter (Fig. 5b). The increase of NO2 −–N was observed in the bottom layer, which suggested that the nitrate addition can have an influence on deeper layers. However, the influence weakened with depth.

Similar trends were observed for NH4 +–N, which decreased over time, in both treatments and control (Fig. 5c). This might be related to the improvement of redox condition in the sediment.

According to Camargo and Alonso (2006), high levels of NH4 +–N, NO2 −–N, and NO3 −–N can impair the ability of aquatic animals to survive, grow, and reproduce, resulting from the toxic effects of the nitrogenous compounds. Our data indicated that calcium nitrate addition temporally increased the concentration of NO2 −–N and NO3 −–N but quickly recovered while Phoslock® addition had negligible effects. Therefore, both treatments are unlikely to cause nitrogenous associated risks.

4 Conclusions

Calcium nitrate application temporally decreased dissolved oxygen in overlaying water but improved the redox condition in sediment in a long run. Whereas Phoslock® only slightly increased the DO in overlaying water. Both treatments reduced SRP and TP in water but their concentrations returned back overtime in calcium nitrate treatment while stayed low in Phoslock® treatment. Phosphorus fractionation in the sediment profile suggested that calcium nitrate addition increased mobile phosphorus fractions (NH4Cl-P and BD-P), which are potentially bioavailable. Whereas Phoslock® addition increased HCl-P, which is considered non-bioavailable and have a low release potential. Calcium nitrate increased the rate of autotrophic denitrification, causing a short-term elevation of nitrite in water and sediment and ammonia in water. The added nitrate was consumed rapidly overtime with only 2 % left at the end of the experiment, likely due to the denitrification process. Nitrite and ammonia in water and nitrite in sediment also declined remarkably following the consumption of nitrate. Phoslock® showed minor influence on nitrogen species. Our results suggested that Phoslock® is a better choice for phosphorus control in eutrophicated lakes, while the effectiveness of calcium nitrate can be more sensitive to environmental conditions. However, calcium nitrate can improve redox condition of the sediment and stimulate indigenous microbial activities, which can provide other benefits for remediation. Therefore, calcium nitrate might be used together with Phoslock® for a better in situ treatment result.