Abstract

High-resolution combined ZrO-Chelex diffusive gradients in thin film (DGT) probes were arranged to synchronously monitor the characteristics of labile phosphorus (P), iron (Fe), and manganese (Mn) in sediments from regions of varying trophic levels in Lake Taihu, China. Results showed that P release was closely related to the lake region's trophic level. Labile P in sediments and the apparent diffusion flux (Fd) of P were higher in hyper-eutrophic algae-dominated regions than those in macrophyte-algae-dominated and macrophyte-dominated regions, indicating that P released from sediments may be facilitated in eutrophic environments. A positive correlation observed between DGT-labile P and Fe or Mn, which had similar regional spatial distributions to P, showed that reductive dissolution of Fe-Mn (oxyhydr)oxides (FeOOH-MnOOH) in sediments was responsible for internal P release, which was strengthened in eutrophic environments. Ratios of labile Fe or Mn/labile P (LFe/LP and LMn/LP) in sediments in algae-dominated regions showed lower values than those in macrophyte-algae-dominated and macrophyte-dominated regions, further confirming that FeOOH-MnOOH in algae-dominated regions were less able to fix P, which brought higher pollution risks for the overlying water. LFe/LP or LMn/LP also had a significantly positive correlation with Fd, further testament that labile Fe or Mn was much more relevant to the sediment P cycle.

INTRODUCTION

Phosphorus (P) accumulation and release in fresh water have been studied extensively in recent decades. Excessive P release caused by human activities leads to widespread eutrophication and algal blooms (Schindler 2006). Once P enters the freshwater system, it is likely to accumulate in sediments through deposition and adsorption, resulting in high sediment P loading (Hupfer et al. 2007). In terms of lake management, even if external P sources are effectively controlled, the internal release of P accumulated in the sediment also contributes to continuous eutrophication (Søndergaard et al. 2003). At present, it is widely believed that the oxidation-reduction of iron and manganese (oxyhydr)oxides (FeOOH-MnOOH) plays an important role in sediment P cycling (Lovley 1991; Smith et al. 2011). Under reductive conditions, FeOOH-MnOOH is reduced to Fe2+ and Mn2+, which results in P release (coprecipitated with or adsorbed on to FeOOH-MnOOH) to pore waters and even to overlying waters. Under oxidative conditions, Fe2+ and Mn2+ are reoxidized to FeOOH-MnOOH, which can adsorb labile P and form a precipitate. Therefore, sediment P participates in early diagenetic processes in sediments through oxidation-reduction of FeOOH-MnOOH. As lake sediments have high spatial heterogeneity, composition is liable to vary, and thus, acquiring spatial distribution information of Fe, Mn, and P in high resolution was very important for detailed studies of P release in sediment.

Diffusive gradients in thin films (DGT) is a passive sampling technique capable of tracking bioavailable fractions of elements in sediment, water, and soil (Zhang et al. 2014). DGT technology not only reflects labile elements in dynamic equilibrium between the solid and liquid phases, but also helps to simplify complicated operations and to minimize analytical errors that occurr in traditional chemical extraction methods (Zhang et al. 2014). Furthermore, DGT technology has higher resolution, shorter balance time, and better selectivity than the techniques of dialysis and membrane diffusion equilibrium (Stockdale et al. 2008). At present, DGT technology has been successfully used to analyze labile P in sediment profiles (Zhang et al. 2014) and the spatial distribution of labile Fe and Mn (Gao et al. 2017). The latest ZrO-Chelex binding DGT technique (ZrO-Chelex DGT) can synchronously obtain information about labile P, Fe, and Mn in sediments at the millimeter scale (Wang et al. 2016) and, thus, has advantages in monitoring the mobility of these ions at high resolution as well as the ions’ coupling relationships.

Sediments from varying trophic regions in Lake Taihu (Wuxi, China) have different P adsorption capacities (Xiao et al. 2015). When the nutrient adsorption quantity exceeds the sediments' load, sediment P can be released into the water column under certain conditions, enhancing the internal pollution risk. In this study, sediments from four regions in Lake Taihu with varying trophic levels, namely Zhushan Bay (ZB), Meiliang Bay (MB), Gonghu Bay (GB), and Southern Taihu (ST), were chosen to study the distribution characteristics of labile P, Fe and Mn, and their relationships, through synchronously monitoring the high-resolution information of labile P, Fe, and Mn in sediment profiles. Meanwhile, the apparent diffusion flux (Fd) of P in the sediment-water interface (SWI) was also calculated to quantitatively assess P mobility in sediments from these regions. This study provides a multiple-perspective approach to systematically investigating labile P cycling coupled with labile Fe and Mn cycling at the SWI of Lake Taihu. It will also contribute to a better understanding of internal reductive release of labile P in eutrophic environments.

MATERIALS AND METHODS

Study site

Two sampling sites were selected in each region of the lake, including ZB, MB, GB and ST (Figure 1). ZB and MB belong to hyper-eutrophic algae-dominated regions, GB is regarded as a middle eutrophic macrophyte-algae-dominated region, while ST is classified as a mesotrophic macrophyte-dominated region.

Figure 1

Distribution of sampling sites in Lake Taihu.

Figure 1

Distribution of sampling sites in Lake Taihu.

Preparation of DGT experiment

DGT probes equipped with ZrO-Chelex gel as a binding layer were used for simultaneous measurements of labile P, Mn, and Fe in sediments (Xu et al. 2013; Wang et al. 2016). Flat probes with 18 × 160 mm (width × length) exposure area were used for measurements. During the assembly of DGT, the ZrO-Chelex binding gel was laid sequentially as a 0.4-mm-thick diffusive gel and a 0.13-mm-thick cellulose nitrate filter membrane (0.45-mm pore size, Whatman, UK), covered with a plastic plate and tied with fishing line (justron, 0.33 mm in diameter). The ZrO-Chelex DGT probes were deoxygenated with nitrogen overnight and stored in a container of deoxygenated sodium chloride solution (0.1 M) before use.

Sampling and processing

In May 2015, water samples were obtained in triplicate at each sampling site. 50 mL of the samples was filtered through a 0.45 μm cellulose acetate membrane and acidified to a pH level of 3 or 4 with HNO3 for Fe and Mn analyses. All samples were stored at 4 °C and measured within one week of collection. For analysis of total nitrogen (TN), total phosphorus (TP), total dissolved phosphorus (TDP) and dissolved reactive phosphorus (DRP), 500 mL of samples was collected, stored at 4 °C, and measured within a 24 h our period. Water parameters in situ, including pH and dissolved oxygen (DO), were measured using a multiple water quality analyzer (YSI 6600v2).

Sediment cores of 80 cm in diameter were collected from each site at 15 cm of overlying water and 25 cm sediment depth. All samples were sealed, stored under darkened conditions, and immediately transported to the laboratory for DGT measurements.

Experiment and sample analysis

All cores were kept in an incubator under darkened conditions at a temperature equivalent to the sampling temperature (∼20 °C). Each DGT probe was plotted to display the SWI position, then was gently inserted in the sediment core and left for 24 hours. After retrieval, probes were rinsed with lake water and then deionized water to remove visible sediment particles.

The ZrO-Chelex gels were disassembled from the DGT probes and vertically cut into two slices to minimize the possible heterogeneity of the sediment cores, from 0 to 9 mm and from 9 to 18 mm. A 5 mm slice of each vertical binding gel was cut and put into a 1.5 ml centrifuge tube. Each slice was then eluted using 1.00 mL of 1.0 M HNO3 for 16 hours for Fe and Mn extraction, washed using deoxygenated water, and then eluted using 1.00 mL of 1.0 M NaOH for 16 hours for P extraction.

TP, DTP, and DRP of in-situ water samples as well as eluted P were measured using the molybdenum blue method (Murphy & Riley 1962). TN of water samples was determined after digestion in alkaline potassium persulfate, which was followed by UV spectrophotometric analysis (Thermo Evolution 300). The Mn of in-situ water samples as well as the Mn in DGT elutes were determined by inductively coupled plasma mass spectrometry (ICP-MS, Agilent 7500cx). Fe concentration of in-situ water samples and DGT elutes were measured by the phenanthroline colorimetric method (Tamura et al. 1974). All water samples were analyzed in triplicate.

The surface sediment was used for chemical analysis. Sediments were freeze-dried at −80 °C and passed through a 0.2-mm sieve for further characterization and analysis. Sediment pH was measured in a 1:2.5 solid (the liquid ratio suspension was determined using a pH electrode). Total organic carbon (TOC) content in sediment samples was analyzed using a TOC analyzer (TOC-Vcph, Elementar, Germany). Total amounts of P, Fe and Mn were determined by X-ray fluorescence (XRF) spectroscopy (AxiosmAX, PANalytical). All analyses were conducted in triplicate. Weakly absorbed and strongly absorbed Fe or Mn were extracted, respectively (Keon et al. 2001; Seddique et al. 2008), and the sum of weakly absorbed and strongly absorbed Fe or Mn was defined as FeOOH or MnOOH. For each step of the extraction procedure, calibration solutions were prepared in the same matrix.

Data analysis and statistics

The concentrations of labile P, Fe, and Mn obtained by DGT were calculated using the following equation:  
formula
(1)
where Δg is the thickness of the diffusive layer (cm), D is the diffusion coefficient of the target elements in the diffusive layer (cm2), A is the exposure window area (cm2), and t is the deployment time (s). M is the accumulated mass of fractions over the deployment time (μg) and varies with the fixed phases and elution agents, so the elution rate needs to be considered as:  
formula
(2)
where Ce is the elution reagent concentration, Ve is the elution reagent volume, Vg is the combination phase volume, usually negligible, and fe is the elution factor.
The Fd of P across the SWI was quantified from labile P profiles using Equation (3) (Xu et al. 2013). Given the different resupply mechanisms of DGT measurements from the overlying water and sediments, the apparent flux was the sum of the sediment and overlying water flux.  
formula
(3)

where Jw and Js represent the labile P flux from sediments to the SWI and from water to the SWI, respectively. It is important to note that the focus of this study was the area in the vicinity of the SWI. For Equation (3), φ is the sediment porosity, for which a value of 0.9 was selected, based on the results of a study by Gao et al. (Gao et al. 2009); Dw and Ds are the P diffusion coefficients in water and sediment, respectively (Xu et al. 2013); and (δCDGT/δXw)(x=0) and (δCDGT/δXs)(x=0) represent labile concentration gradients in the overlying water and sediment, respectively. A distance of approximately 2 cm from the SWI within the experimental area was selected and gradients were determined within this area.

The spatial differences of each parameter in water and sediment were analyzed by variance analysis (ANOVA) using post-hoc single comparisons (Duncan's new multiple range test (MRT)). The correlation coefficients between P indicators in sediment and the main geochemical parameters were calculated to assess the influence of the main geochemical parameters on P release. Both statistical analyses were computed using SPSS 19.0 software.

RESULTS AND DISCUSSION

Basic properties of overlying water and sediments

TN, TP, TDP, and DRP content of the overlying water showed spatial differences (p < 0.05, Table 1). The highest concentrations of TN, TP, TDP, and DRP occurred in ZB with corresponding values of 3.91, 0.16, 0.14, and 0.12 mg L−1, respectively. The pH of overlying water was generally weakly alkaline (7.37–8.37). The DO mean values ranged from 6.80 to 9.53 mg L−1, showing that the overlying water of each lake region was aerobic in nature. DO content of ZB was the lowest, which possibly resulted from the intense oxygen decomposition by a large number of phytoplankton in this lake region (Yan et al. 2016). At the same time, concentrations of dissolved Fe and Mn in overlying water yielded higher values in this lake region (113.05 and 38.02 mg L−1), indicating that the region was in a stronger reductive environment.

Table 1

Physicochemical properties of the overlying water and surface sediments of Lake Taihu

    Zhushan Bay
 
Meiliang Bay
 
Gonghu Bay
 
Southern Taihu
 
ZB-1 ZB-2 Average MB-1 MB-2 Average GB-1 GB-2 Average ST-1 ST-2 Average 
Sediment Depth (m) 1.88 2.67 2.27 2.61 2.65 2.63 2.33 2.56 2.44 2.25 2.38 2.31 
pH 7.48 7.36 7.42 7.92 8.19 8.05 7.82 7.49 7.65 8.11 8.22 8.16 
TOC (%) 1.53 1.39 1.46a 0.90 1.29 1.09a,b 0.94 0.66 0.80b 0.77 0.74 0.75b 
TP (g kg−11.32 0.96 1.14a 0.47 0.89 0.68b 0.34 0.44 0.39b 0.56 0.53 0.55b 
Fe (g kg−147.99 40.34 44.17 34.11 35.59 37.92 35.09 29.22 34.85 38.47 37.36 32.16 
Mn (g kg−12.68 1.79 2.24a 0.61 0.75 0.68b 0.73 0.47 0.60b 1.26 1.01 1.13b 
FeOOH (g kg−123.16 21.06 22.11a 20.80 19.51 20.15a,b 18.32 15.71 17.01b 17.16 16.79 16.97b 
MnOOH (g kg−11.44 1.01 1.22a 0.54 0.80 0.67b 0.32 0.26 0.29b 0.55 0.52 0.53b 
Overlying water pH 8.25 8.16 8.20a,b 8.04 7.85 7.94b,c 7.37 7.56 7.46c 8.37 8.31 8.34a 
DO (mg L−16.49 7.11 6.80b 9.19 9.25 9.22a 9.46 9.14 9.30a 9.56 9.50 9.53a 
TN (mg L−14.15 3.68 3.91a 3.53 3.47 3.50a,b 3.44 3.21 3.32a,b 3.38 2.92 3.15b 
TP (mg L−10.18 0.14 0.16a 0.091 0.087 0.089b 0.062 0.070 0.066b 0.093 0.081 0.087b 
TDP (mg L−10.16 0.13 0.14a 0.028 0.027 0.028b 0.016 0.026 0.021b 0.017 0.019 0.018b 
DRP (mg L−10.13 0.11 0.12a 0.015 0.014 0.015b 0.006 0.014 0.010b 0.014 0.014 0.014b 
Fe (mg L−1112.19 113.92 113.05a 98.52 96.53 97.52a,b 103.36 98.51 100.93a,b 74.72 69.76 72.24b 
Mn (mg L−138.45 37.59 38.02a 17.30 12.77 15.03b 15.22 12.27 13.74b 19.95 16.66 18.30b 
    Zhushan Bay
 
Meiliang Bay
 
Gonghu Bay
 
Southern Taihu
 
ZB-1 ZB-2 Average MB-1 MB-2 Average GB-1 GB-2 Average ST-1 ST-2 Average 
Sediment Depth (m) 1.88 2.67 2.27 2.61 2.65 2.63 2.33 2.56 2.44 2.25 2.38 2.31 
pH 7.48 7.36 7.42 7.92 8.19 8.05 7.82 7.49 7.65 8.11 8.22 8.16 
TOC (%) 1.53 1.39 1.46a 0.90 1.29 1.09a,b 0.94 0.66 0.80b 0.77 0.74 0.75b 
TP (g kg−11.32 0.96 1.14a 0.47 0.89 0.68b 0.34 0.44 0.39b 0.56 0.53 0.55b 
Fe (g kg−147.99 40.34 44.17 34.11 35.59 37.92 35.09 29.22 34.85 38.47 37.36 32.16 
Mn (g kg−12.68 1.79 2.24a 0.61 0.75 0.68b 0.73 0.47 0.60b 1.26 1.01 1.13b 
FeOOH (g kg−123.16 21.06 22.11a 20.80 19.51 20.15a,b 18.32 15.71 17.01b 17.16 16.79 16.97b 
MnOOH (g kg−11.44 1.01 1.22a 0.54 0.80 0.67b 0.32 0.26 0.29b 0.55 0.52 0.53b 
Overlying water pH 8.25 8.16 8.20a,b 8.04 7.85 7.94b,c 7.37 7.56 7.46c 8.37 8.31 8.34a 
DO (mg L−16.49 7.11 6.80b 9.19 9.25 9.22a 9.46 9.14 9.30a 9.56 9.50 9.53a 
TN (mg L−14.15 3.68 3.91a 3.53 3.47 3.50a,b 3.44 3.21 3.32a,b 3.38 2.92 3.15b 
TP (mg L−10.18 0.14 0.16a 0.091 0.087 0.089b 0.062 0.070 0.066b 0.093 0.081 0.087b 
TDP (mg L−10.16 0.13 0.14a 0.028 0.027 0.028b 0.016 0.026 0.021b 0.017 0.019 0.018b 
DRP (mg L−10.13 0.11 0.12a 0.015 0.014 0.015b 0.006 0.014 0.010b 0.014 0.014 0.014b 
Fe (mg L−1112.19 113.92 113.05a 98.52 96.53 97.52a,b 103.36 98.51 100.93a,b 74.72 69.76 72.24b 
Mn (mg L−138.45 37.59 38.02a 17.30 12.77 15.03b 15.22 12.27 13.74b 19.95 16.66 18.30b 

Different superscripts indicate significant differences in the parameters among the four regions (p < 0.05) according to one-way ANOVA. a, b, and c represent significantly different classification of parameters among regions in descending order.

The pH values of the sediments were similar to that of the overlying water, i.e. also weakly alkaline (7.36–8.22) (Table 1). The average TOC content was in the range of 0.75–1.46%, and the highest TOC content was observed in algae-dominated lake regions (ZB and MB, 1.46% and 1.09%, respectively), indicating that long-term accumulation of organic matter in eutrophic sediments was extremely significant in such regions. The mean contents of FeOOH and MnOOH were in the range of 16.97–22.11 mg kg−1 and 0.29–1.22 mg kg−1, respectively. The highest FeOOH and MnOOH were also found in sediments from algae-dominated lake regions (ZB and MB). The order of decreasing average TP in investigated sediments ranked as follows: ZB (1.14 g kg−1) > MB (0.68 g kg−1) > ST (0.55 g kg−1) > GB (0.39 g kg−1) (Table 1). The sediment TP distribution was in line with that of the overlying water. Moreover, TP in sediments was also correlated with FeOOH (r = 0.776; p < 0.05) and MnOOH (r = 0.981; p < 0.01), demonstrating that the cycling of Fe and Mn may play an important role in P sedimentary partitioning.

Distribution characteristics of DGT-labile P, Fe, and Mn

The vertical and transversal distributions of DGT-labile P, Fe, and Mn in the sediments showed that labile P exhibited varying profile trends in different lake regions (Figure 2). The vertical distribution of labile P in sediments from these regions can generally be classified into three patterns. In algae-dominated lake regions (ZB and MB), the labile P in sediments increased with depth. Although the labile P in site MB-1 increased slowly in the deep zone, it was still different from the labile P profiles obtained in GB and ST regions. Labile P in macrophyte-algae-dominated lake regions (GB) increased with depth and had a peak in the middle zone (∼40–80 mm), and then decreased to the bottom. Labile P in sediments from ST were low and increased slightly with depth. These three profile distributions of labile P were basically consistent with the reported literature (Ding et al. 2015).

Figure 2

The distribution of available P, Fe and Mn in the vertical and transverse sections of sediment.

Figure 2

The distribution of available P, Fe and Mn in the vertical and transverse sections of sediment.

Labile Fe and Mn in ZB sediments showed similar trends to the labile P in the whole profile (Figure 2(b)). For the other sites, the labile P, Fe, and Mn showed similar trends within the depth of 0–70 mm; however, dissimilar trends occurred below this depth range. Inconsistent trends of labile Fe, Mn, and P in the deep zone might be caused by different diagenesis processes. In GB and ST, the nonsynchronous trends of labile Fe, Mn, and P in the deep zone may be for two reasons: the first reason being that sedimentary diagenesis intensity increased with depth, P adsorbed on mineral particles can be converted into more stable phosphate minerals, leading to the decrease of labile P (Eckert & Nishri 2014). Another possible explanation for decreasing labile P in the deep zone was that, in the macrophyte-algae-dominated and macrophyte-dominated regions, the growth of aquatic plants created a great demand for P from both water and sediment. In MB, Fe-bound P was more abundant in the deep zone and was easier to dissolve in the strongly reductive environment (Yuan et al. 2010), which induced the almost consistent trends of labile Fe and P. Unlike labile Fe, labile Mn in the deep zone was not in step with labile P. Such inconsistent trends likely resulted from the lower MnOOH content in this region (Table 1), which released low labile Mn to react easily with Ca to form Ca-Mn carbonate precipitates (Haese et al. 2000). Therefore, when the reductive dissolution of FeOOH and MnOOH in deep sediments was the dominant factor, the tendencies of labile Fe, Mn, and P in the deep zone moved towards synchronization. When the precipitation or adsorption process played a dominant role, the three trends tended towards non-synchronization.

The average labile Fe, Mn, and P in sediments were calculated to study their spatial differences (Table 2). The mean values of labile P, Fe, and Mn in ZB were 23, 2, and 2 times that of ST, indicating that the algae-dominated region was in a strongly reductive environment which brought about a significant reduction of FeOOH and MnOOH to release P. The significant positive correlation between the labile P content and labile Fe or Mn further indicated that P release was subject to the cycles of Fe and Mn (Smith et al. 2011) (Table 3). At the same time, the correlation analysis between P indicators in sediments and those in the overlying water showed that labile P had a more significant correlation with DRP and DTP than TP in sediments. In particular, at the depth of 0–20 mm, labile P had the most significant correlation with various P species in the overlying water (Table 2). Thus, the labile P can better reflect the activity of sediment P. The significant correlation between P flux and DRP or DTP further supported the above conclusion. Furthermore, both labile P and TPsediment were correlated with TOC (Table 2), indicating that organic matter played an important role in P mobility. TOC promoted the reduction and decomposition of iron oxide, thereby promoting P mobility (Stockdale et al. 2009). At the same time, the oxidation of organic matter and the accompanying reduction of associated oxidation products could further enhance sediment P release (Withers & Jarvie 2008). Horizontal results indicated that labile P showed significant differences but similar variation tendencies between the two slice samples. This result is evidence that high-resolution measurements greatly facilitate the discovery of small variations caused by sediment heterogeneity. Meanwhile, such vertical analyses could also minimize geological diversity between parallel sites.

Table 2

Correlation between DGT-labile P and the each indicator of sediments

Sediment TP, overlying water TDP DRP DGT-labile Mn, sediment (0–120 mm) DGT-labile Fe, sediment (0–120 mm) TOC 
TP 0.916** 0.878** 0.876** 0.839** 0.646 0.900** 
DGT-labile P, sediment (0–20 mm) 0.923** 0.909** 0.895** 0.868** 0.869** 0.807* 
DGT-labile P, sediment (0–120 mm) 0.920** 0.898** 0.883** 0.860** 0.850** 0.845** 
Fd 0.713* 0.759* 0.727* 0.602 0.721* 0.820* 
Sediment TP, overlying water TDP DRP DGT-labile Mn, sediment (0–120 mm) DGT-labile Fe, sediment (0–120 mm) TOC 
TP 0.916** 0.878** 0.876** 0.839** 0.646 0.900** 
DGT-labile P, sediment (0–20 mm) 0.923** 0.909** 0.895** 0.868** 0.869** 0.807* 
DGT-labile P, sediment (0–120 mm) 0.920** 0.898** 0.883** 0.860** 0.850** 0.845** 
Fd 0.713* 0.759* 0.727* 0.602 0.721* 0.820* 

*p < 0.05:**p < 0.01 (n = 8).

Table 3

The average values of DGT-labile Fe, Mn and P and the ratios in each sediment

Site DGT-labile Fe DGT-labile Mn DGT-labile P LFe/LLMn/L
ZB-1 245.67 253.52 247.55 0.75 1.25 
ZB-2 177.36 202.51 102.37 1.43 1.77 
MB-1 196.19 106.91 74.91 3.34 1.26 
MB-2 149.22 110.05 69.36 2.62 4.26 
GB-1 121.76 104.74 14.06 16.05 9.49 
GB-2 143.42 84.59 16.95 10.32 7.79 
ST-1 64.17 72.33 6.83 8.54 12.87 
ST-2 144.51 146.81 8.28 19.70 27.55 
Site DGT-labile Fe DGT-labile Mn DGT-labile P LFe/LLMn/L
ZB-1 245.67 253.52 247.55 0.75 1.25 
ZB-2 177.36 202.51 102.37 1.43 1.77 
MB-1 196.19 106.91 74.91 3.34 1.26 
MB-2 149.22 110.05 69.36 2.62 4.26 
GB-1 121.76 104.74 14.06 16.05 9.49 
GB-2 143.42 84.59 16.95 10.32 7.79 
ST-1 64.17 72.33 6.83 8.54 12.87 
ST-2 144.51 146.81 8.28 19.70 27.55 

Each value represents the mean value of the two transverse profiles in each site (0–9 mm, 9–18 mm).

Estimation of apparent diffusion flux of P across the SWI

The apparent diffusion flux of P was calculated from the high-resolution P distribution across the SWI. As shown in Figure 3, the average Fd of P was in the range of –6.17 to 17.80 ng cm−2 d−1, and Fd in algae-dominated regions (ZB-1, ZB-2, MB-1, and MB-2: 17.80, 13.40, 10.83, and 8.31 ng cm−2 d−1, respectively) were far above those measured in macrophyte-algae-dominated and macrophyte-dominated regions (GB-1, GB-2, ST-1, and ST-2: 6.33, 3.69, 0.78, and –6.17 ng cm−2 d−1, respectively). These results showed that labile P in sediments from algae-dominated regions had the stronger release potential to the overlying water, which made the sediments the internal pollution source in Lake Taihu, thus having an important impact on the lake's eutrophication. In contrast, the sediments in macrophyte-algae-dominated and macrophyte-dominated regions displayed low P release potential or existed as a P repository, buffering regional P pollution. This result was in accordance with previous reports that internal P loading in the polluted or algae-dominated lake region was stronger than that in macrophyte-algae-dominated or macrophyte-dominated lake regions (Kisand & Nõges 2003; Zhu et al. 2013).

Figure 3

The apparent diffusion flux of P across the SWI of each sampling site.

Figure 3

The apparent diffusion flux of P across the SWI of each sampling site.

Comparison of the ability of FeOOH-MnOOH fixing labile P in sediments

The ability of FeOOH-MnOOH to fix P was assessed by calculation of the ratio between labile Fe or Mn and labile P (defined as LFe/LP and LMn/LP, respectively), and profiles of the average ratios of LFe/LP and LMn/LP are shown in Figure 4. For sites in macrophyte-algae-dominated or macrophyte-dominated lake regions (GB-1, GB-2, ST-1 and ST-2), both LFe/LP and LMn/LP increased gently or sharply from SWI to the middle zone, then decreased with depth. The distribution features indicated that FeOOH and MnOOH of these surface sediments possessed strong P-fixation ability. For sites in algae-dominated region (MB-1, MB-2, ZB-1, and ZB-2), LFe/LP or LMn/LP increased only slightly with depth, suggesting in these sites both FeOOH and MnOOH showed a relatively weak binding capacity to P, facilitating P release to overlying water. Small peaks of LFe/LP or LMn/LP in MB-1 and MB-2 were likely caused by geological heterogeneity. For sites in macrophyte-algae-dominated or macrophyte-dominated lake regions, the values of LFe/LP and LMn/LP were in the range of 8.54–19.70 and 7.79–27.55, respectively. The corresponding values in sites in algae-dominated lake regions were in the range of 0.75–3.34 and 1.25–4.26, respectively (Table 3). The lower LFe/LP and LMn/LP suggested that P in sediments from algae-dominated lake regions had stronger potential to release from FeOOH-MnOOH, which brought higher pollution risk to the overlying water.

Figure 4

The average distributions of LFe/LP and LMn/LP in sediment profiles.

Figure 4

The average distributions of LFe/LP and LMn/LP in sediment profiles.

The linear correlation slope between the labile Fe or Mn and the labile P was calculated to further examine the ability of FeOOH-MnOOH to control the internal P release (Table 4). The slopes of LFe/LP and LMn/LP in sites in the algae-dominated lake regions (1.21–2.24 and 0.35–2.28) were smaller than those in macrophyte-algae-dominated or macrophyte-dominated lake regions (3.75–15.60 and 1.82–7.59). The results further indicated that FeOOH or MnOOH in sediments from macrophyte-algae-dominated or macrophyte-dominated lake regions were more capable of adsorbing P, which was consistent with the ratio of LFe/LP or LMn/LP, Fd of P and the labile P concentration. Jensen et al. have proposed a total amount ratio (TFe/TP) to assess whether the surface of the oxygenated sediment had sufficient adsorption capacity for P (critical value is 15) (Jensen et al. 1992). Based on this, the ratio of TFe/TP in each site was calculated from Table 1, and the values in ZB, MB, GB and ST were found to be 39.08, 55.97, 84.74 and 69.53, respectively. All these ratios were higher than the critical value, suggesting that FeOOH in all lake sediments had strong P-fixation ability. However, the results were not consistent with the results of Fd, which showed that in algae-dominated regions the sediments' adsorption capacity of fixing P was weak. The TFe/TP ratio which was used to assess the binding capacity of FeOOH to P therefore has its limitations. In addition, the labile ratio slope had a significant negative correlation with Fd, while the total amount ratio of TFe/TP and TMn/TP showed no correlation with Fd (Table 5), which further illustrated that labile Fe and Mn were more closely correlated to the cycle of P in sediments than the total Fe and Mn. In summary, compared with the total amount, the labile content can provide a more accurate assessment in determining Fe-coupled and Mn-coupled P release in sediments. Besides, as deep sediments may also be affected by other processes, a more comprehensive study is required to reveal micro-change processes associated with P in these regions.

Table 4

Correlation between the DGT-labile Fe or Mn and DGT-labile P in profiles of sediments (n = 50)

Site Linear equation between DGT-labile Fe and P Linear equation between DGT-labile Mn and P 
ZB-1 LFe = 1.29 LP–72.69 (r = 0.960, p < 0.01) LMn = 0.90 LP + 30.06 (r = 0.806, p < 0.01) 
ZB-2 LFe = 2.25 LP–52.62 (r = 0.868, p < 0.01) LMn = 2.28 LP–30.87 (r = 0.899, p < 0.01) 
MB-1 LFe = 1.22 LP + 104.99 (r = 0.560, p < 0.01) LMn = 1.64 LP–15.56 (r = 0.844, p < 0.01) 
MB-2 LFe = 1.85 LP + 20.66 (r = 0.929, p < 0.01) LMn = 0.35 LP + 85.70 (r = 0.330, p < 0.05) 
GB-1 LFe = 3.76 LP + 101.53 (r = 0.377, p < 0.01) LMn = 2.95 LP + 50.03 (r = 0.479, p < 0.01) 
GB-2 LFe = 4.61 LP + 60.48 (r = 0.685, p < 0.01) LMn = 1.82 LP + 51.93 (r = 0.372, p < 0.01) 
ST-1 LFe = 15.60 LP–42.35 (r = 0.724, p < 0.01) LMn = 7.01 LP + 36.28 (r = 0.763, p < 0.01) 
ST-2 LFe = 14.70 LP + 19.72 (r = 0.795, p < 0.01) LMn = 7.60 LP + 64.77 (r = 0.887, p < 0.01) 
Site Linear equation between DGT-labile Fe and P Linear equation between DGT-labile Mn and P 
ZB-1 LFe = 1.29 LP–72.69 (r = 0.960, p < 0.01) LMn = 0.90 LP + 30.06 (r = 0.806, p < 0.01) 
ZB-2 LFe = 2.25 LP–52.62 (r = 0.868, p < 0.01) LMn = 2.28 LP–30.87 (r = 0.899, p < 0.01) 
MB-1 LFe = 1.22 LP + 104.99 (r = 0.560, p < 0.01) LMn = 1.64 LP–15.56 (r = 0.844, p < 0.01) 
MB-2 LFe = 1.85 LP + 20.66 (r = 0.929, p < 0.01) LMn = 0.35 LP + 85.70 (r = 0.330, p < 0.05) 
GB-1 LFe = 3.76 LP + 101.53 (r = 0.377, p < 0.01) LMn = 2.95 LP + 50.03 (r = 0.479, p < 0.01) 
GB-2 LFe = 4.61 LP + 60.48 (r = 0.685, p < 0.01) LMn = 1.82 LP + 51.93 (r = 0.372, p < 0.01) 
ST-1 LFe = 15.60 LP–42.35 (r = 0.724, p < 0.01) LMn = 7.01 LP + 36.28 (r = 0.763, p < 0.01) 
ST-2 LFe = 14.70 LP + 19.72 (r = 0.795, p < 0.01) LMn = 7.60 LP + 64.77 (r = 0.887, p < 0.01) 
Table 5

Correlative analysis between the Fd and the ratios (n = 8)

Linear equation p 
Fd = −1.07 LFe/LP + 12.91 −0.848 <0.01 
Fd = −2.24 LMn/LP + 13.75 −0.814 <0.01 
Fd = −0.17 TFe/TP + 17.422 −0.518 0.188 
Fd = −1.18 TMn/TP + 8.850 −0.079 0.853 
Linear equation p 
Fd = −1.07 LFe/LP + 12.91 −0.848 <0.01 
Fd = −2.24 LMn/LP + 13.75 −0.814 <0.01 
Fd = −0.17 TFe/TP + 17.422 −0.518 0.188 
Fd = −1.18 TMn/TP + 8.850 −0.079 0.853 

CONCLUSION

In this study, the distribution characteristics of labile P, Fe, and Mn in lake sediments were successfully monitored and assessed using a high-resolution ZrO-Chelex DGT technique. The labile P content and the apparent diffusion flux of P in sediments in algae-dominated lake regions were higher than those in the macrophyte-algae-dominated or macrophyte-dominated lake regions. DGT-labile P in each site was positively correlated with the DGT-labile Fe and Mn, indicating that the oxidation-reduction of FeOOH-MnOOH was important to the sediment P cycle. The values of LFe/LP and LMn/LP in sediments in algae-dominated lake regions were lower than those in macrophyte-algae-dominated or macrophyte-dominated lake regions, indicating that P in sediments in algae-dominated lake regions had stronger potential to release from FeOOH-MnOOH, which brought out a higher pollution risk to the overlying water. The slopes of LFe/LP and LMn/LP in algae-dominated lake regions were smaller than those in macrophyte-algae-dominated or macrophyte-dominated lake regions, indicating that FeOOH-MnOOH in algae-macrophyte-dominated or macrophyte-dominated lake regions were more capable of fixing P. The slopes of LFe/LP and LMn/LP had significant negative correlations with Fd, while TFe/TP and TMn/TP had no correlation with Fd, suggesting LFe/LP and LMn/LP provided a more accurate assessment in determining Fe-coupled and Mn-coupled P release in sediments.

ACKNOWLEDGEMENTS

The authors of this study are grateful for the support from the National Natural Science Foundation of China (NSFC) (21577138 and 21277136). We would also like to express special thanks to the various anonymous reviewers for their helpful comments.

REFERENCES

REFERENCES
Ding
,
S.
,
Han
,
C.
,
Wang
,
Y.
,
Yao
,
L.
,
Wang
,
Y.
&
Xu
,
D.
2015
In situ, high-resolution imaging of labile phosphorus in sediments of a large eutrophic lake
.
Water Research
74
,
100
109
.
Eckert
,
W.
&
Nishri
,
A.
2014
The Phosphorus Cycle. Lake Kinneret
.
Springer
,
The Netherlands
, pp.
347
363
.
Gao
,
Y.
,
Lesven
,
L.
,
Gillan
,
D.
,
Sabbe
,
K.
,
Billon
,
G.
,
De Galan
,
S.
,
Galan
,
S.
,
Elskens
,
M.
,
Baeyens
,
W.
&
Leermakers
,
M.
2009
Geochemical behavior of trace elements in sub-tidal marine sediments of the Belgian coast
.
Marine Chemistry
117
(
1–4
),
88
96
.
Gao
,
L.
,
Gao
,
B.
,
Xu
,
D.
,
Peng
,
W.
,
Lu
,
J.
&
Gao
,
J.
2017
Assessing remobilization characteristics of arsenic (As) in tributary sediment cores in the largest reservoir, China
.
Ecotoxicology and Environmental Safety
140
,
48
54
.
Haese
,
R. R.
,
Schramm
,
J.
,
Van Der Loeff
,
M. R.
&
Schulz
,
H. D.
2000
A comparative study of iron and manganese diagenesis in continental slope and deep sea basin sediments off Uruguay (SW Atlantic)
.
International Journal of Earth Sciences
88
(
4
),
619
629
.
Hupfer
,
M.
,
Gloess
,
S.
&
Grossart
,
H. P.
2007
Polyphosphate-accumulating microorganisms in aquatic sediments
.
Aquatic Microbial Ecology
47
(
3
),
299
311
.
Jensen
,
H. S.
,
Kristensen
,
P.
,
Jeppesen
,
E.
&
Skytthe
,
A.
1992
Iron: phosphorus ratio in surface sediment as an indicator of phosphate release from aerobic sediments in shallow lakes
.
Hydrobiologia
235
(
1
),
731
743
.
Keon
,
N. E.
,
Swartz
,
C. H.
,
Brabander
,
D. J.
,
Harvey
,
C.
&
Hemond
,
H. F.
2001
Validation of an arsenic sequential extraction method for evaluating mobility in sediments
.
Environmental Science & Technology
35
(
13
),
2778
2784
.
Lovley
,
D. R.
1991
Dissimilatory Fe (III) and Mn (IV) reduction
.
Microbiological Reviews
55
(
2
),
259
287
.
Murphy
,
J. A. M. E. S.
&
Riley
,
J. P.
1962
A modified single solution method for the determination of phosphate in natural waters
.
Analytica Chimica Acta
27
,
31
36
.
Schindler
,
D. W.
2006
Recent advances in the understanding and management of eutrophication
.
Limnology and Oceanography
51
,
356
363
.
Seddique
,
A. A.
,
Masuda
,
H.
,
Mitamura
,
M.
,
Shinoda
,
K.
,
Yamanaka
,
T.
,
Itai
,
T.
,
Maruoka
,
T.
,
Uesugi
,
K.
,
Ahmed
,
K. M.
&
Biswas
,
D. K.
2008
Arsenic release from biotite into a Holocene groundwater aquifer in Bangladesh
.
Applied Geochemistry
23
(
8
),
2236
2248
.
Søndergaard
,
M.
,
Jensen
,
J. P.
&
Jeppesen
,
E.
2003
Role of sediment and internal loading of phosphorus in shallow lakes
.
Hydrobiologia
506
(
1
),
135
145
.
Withers
,
P. J. A.
&
Jarvie
,
H. P.
2008
Delivery and cycling of phosphorus in rivers: a review
.
Science of the Total Environment
400
(
1
),
379
395
.
Xiao
,
W. J.
,
Cao
,
X. Y.
,
Song
,
C. L.
&
Zhou
,
Y. Y.
2015
Comparison of fractions and sorption behaviors of sediment phosphorus between typical zones with different trophic states in Lake Taihu, China
.
Chinese Journal of Environmental Engineering
9
(
7
),
3525
3530
.
Xu
,
D.
,
Chen
,
Y.
,
Ding
,
S.
,
Sun
,
Q.
,
Wang
,
Y.
&
Zhang
,
C.
2013
Diffusive gradients in thin films technique equipped with a mixed binding gel for simultaneous measurements of dissolved reactive phosphorus and dissolved iron
.
Environmental Science & Technology
47
(
18
),
10477
10484
.
Yan
,
C.
,
Che
,
F.
,
Zeng
,
L.
,
Wang
,
Z.
,
Du
,
M.
,
Wei
,
Q.
,
Wang
,
Z.
,
Wang
,
D.
&
Zhen
,
Z.
2016
Spatial and seasonal changes of arsenic species in Lake Taihu in relation to eutrophication
.
Science of the Total Environment
563
,
496
505
.
Yuan
,
H.
,
Shen
,
J.
&
Liu
,
E.
2010
Analysis of phosphorus forms in different regions of Taihu Lake
.
China Environmental Science
30
(
11
),
1522
1528
.
Zhang
,
C.
,
Ding
,
S.
,
Xu
,
D.
,
Tang
,
Y.
&
Wong
,
M. H.
2014
Bioavailability assessment of phosphorus and metals in soils and sediments: a review of diffusive gradients in thin films (DGT)
.
Environmental Monitoring and Assessment
186
(
11
),
7367
7378
.