Characteristics of Pore Water Dissolved Organic Nitrogen in the Sediment of Erhai Lake Basin,China

Li Zhang,Shengrui Wang,Qingxuan Sun,Han Liu,Wenzhang Li

1 National Engineering Laboratory for Advanced Municipal Wastewater Treatment and Reuse Technology,Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering,Beijing University of Technology,Beijing 100124,China

2 State Key Laboratory of Environmental Criteria and Risk Assessment,Chinese Research Academy of Environmental Sciences,Beijing,100012,China

3 College of Water Sciences,Beijing Normal University,Beijing 100875,China

Keywords Erhai lake Pore water DON Fluorescence regional integration UV absorbance 3DEEM

Abstract The distribution of Erhai Lake sediment pore water DON followed the pattern of middle(1.449 mg/L)>south(0.828 mg/L)>north(0.266 mg/L),with the highest concentrations being found in the least polluted areas. The sediment pore water DON mainly contained hydrophilic substances and substances derived from microbial sources. Most of these were integrated into humic acid-like substances(～78%),which benefited for reducing sediment DON releasing risk. There were more chains in Erhai Lake sediment pore water DON,with a small number of aromatic substituents. The characteristics of DON components were suggested to be related to human activities.

1 Introduction

Nitrogen is an important nutrient,and it is often a limiting factor in aquatic ecosystems(Han and Li,2016). An increase in N content will affect the structure of the aquatic ecosystem and the composition of aquatic organisms (Wang et al., 2016). Therefore, the N level can directly influence the stability of an aquatic ecosystem.The largest pool of fixed nitrogen in most aquatic systems is dissolved organic nitrogen (DON)(Berman et al.,2003). Research has shown that DON is composed of a series of compounds with nitrogen-containing functional groups, and zero to 73% of the DON from natural and anthropogenic sources can be used by biological communities (Su et al., 2015). Nitrogen is arguably the most important nutrient in regulating primary productivity and species diversity in both aquatic and terrestrial ecosystems(Yang et al.,2016). Moreover,either directly or indirectly, some DON compounds are the main N source supporting algae growth; the DON concentration and the contribution of DON to total dissolved nitrogen have been found to greatly increase during blooms (Bronk et al.,2007;Zhang et al.,2015).

Some DON comes from endogenous sources. Concentration differences between the pore water and the overlying water promote DON diffusive and biologically mediated fluxes across the sediment-water interface(Angela et al., 2000). Porewater, as a medium connecting sediments and overlying water, is important for the identification of biogeochemical processes in aquatic systems. It provides information on organic matter processes such as polymerization or degradation.

Lake sediment can provide continuous time-series and high-resolution information on aquatic ecosystem nutrients (Jing et al., 2007), and it can be used to understand changes in an aquatic ecosystem, such as Erhai Lake. Dissolved organic matter (DOM) plays a major role in many biogeochemical processes in estuarine sediments(David et al.,2001). There is considerable evidence that long residence time facilitates the detection of any allochthonous and autochthonous DOM that accumulates overlong time scales in lakes; this DOM can record the processes of environmental and climate change(Goldberget al.,2015). Humic substances form from the remains of organisms populated a lake and its catchment(Aina et al.,2014),from which we can acquire the traces of lake evolution. As the important component of DOM,DON participates in the turnover of DOM,and it can also be used to understand changes in a lake environment.

Erhai Lake is the second largest freshwater lake in Yunnan Province,which is located in the southwest region of China,playing a crucial role in local socioeconomic development. Therefore,this study made an attempt to:(1)reveal the compositional characteristics and distribution of the sediment pore water DON in Erhai Lake;(2)understand changes in the Lake’s aquatic biochemistry over time. The obtained results could provide support for the prevention and pollution control of Erhai Lake.

2 Materials and methods

2.1 Sampling sites

This study was conducted in Erhai Lake (100.096-100.289°E,25.602-25.964°N).It has an average elevation of 1974 m and a surface area of 249km2(Zhang et al.,2016). At present,Erhai Lake is in the preliminary stages of eutrophication. A significant change had occurred during recent agricultural intensification (Ni et al.,2016).Before the 1970s, Erhai Lake was controlled by natural processes; after the 1970s, the rapid development of agriculture promoted the accumulation of external nutrient pollution load in the lake.

The overlying water DON of Erhai Lake is approximately 0.2 mg/L and the sediment DON is approximately 37.19 mg/kg(Zhang et al.,2016). The average DON/TN(total nitrogen)ratio was relatively higher than that of in other lakes(e.g.,the DON/TN in Chaohu Lake was approximately 20%lower)(Zhang et al.,2016). Ten sites were selected for researching sediment pore water DON based on topographical features of Erhai Lake(Fig. 1).Samples were distributed in three regions: north(EH19,EH21,EH46,and EH73),middle(EH93,EH105,and EH117),and south(EH132,EH142,and EH209)(marked by triangles in Fig. 1).

2.2 Sampling procedure and measurement of water quality parameters

Sediment samples were collected in the prime season for algae blooms. Eight surficial sedimentary samples(0-10 cm)were taken with a Peterson grabsampler (Zhang et al., 2013). Three sedimentary cores, of approximately 30 cm depth, were collected (from EH73(north), EH105(middle), and EH142(south)). Sampling was performed using a gravity sediment corer with a length of 30 cm and an internal diameter of 5 cm. Sediment cores were immediately divided into eight layers: the surficial layer of 0-2 cm,interim layers every 4 cm,and the bottom layer of 26-30 cm. All the samples were packed in plastic bags and stored at 4°C in darkness prior to extracting the pore water within 24h. Water quality parameters, including total nitrogen, dissolved total nitrogen(DTN),and dissolved inorganic nitrogen(NH4-N,NO3-N,and NO2-N)were measured according to the methods previously detailed by Zhang et al. (2016). The inorganic nitrogen concentration in the pore water in each sample of this study were shown in Table S1.

2.3 Analytical methods

2.3.1 Analysis of sediment enzyme activity and microorganism numbers

Sediment enzyme activities were determined as described by Guan(1986). Each enzyme activity was assayed in triplicate, and the averaged data were utilized. Invertase activity was measured by a colorimetric method using 3,5-dinitrosalicylic acid at a wavelength of 500 nm and was expressed in EUinv,where 1 unit is the amount of enzyme that produces 1 milligram glucose per gram of sediment in 1 day at 37°C.Urease activity was measured by the indophenol blue colorimetric method at a wavelength of 460 nm and was expressed in EUura, where 1 unit is the amount of enzyme that produces 1 milligram NH4-N per gram of sediment in 2 days at 37°C.Alkaline phosphatase activity(APA)was measured using a colorimetric method at a wavelength of 410 nm and was expressed in EUapa,where 1 unit is the amount of enzyme that produces 1 mol phenol per gram of sediment in 1 h at 37°C.Protease activity was measured using a colorimetric method at a wavelength of 500 nm and was expressed in EUpro,where 1 unit is the amount of enzyme that produces 1 milligram amino nitrogen per gram of sediment in 1 day at 30°C.Peroxidase activity and polyphenoloxidase activity were measured by a colorimetric method at a wavelength of 430 nm and were expressed in EUplo and EUpol, respectively, where 1 unit is the amount of enzyme that produces 1 mg nutgall per kilogram of sediment in 2 h at 30 ?C.

The number of viable bacteria, fungi and actinomycetes was estimated using the plate dilution technique.The specific procedure was introduced in detail in Zhang et al. (2015).

2.3.2 Ultraviolet-visible absorbance

UV-visible spectra using a 1 cm quartz cuvette on a Hach DR-5000 spectrophotometer (Hach, USA)at wavelengths ranging from 200 to 700 nm was applied for analysis here, with Milli-Q water as a blank. Scan mode was used to measure the absorbance from 200 nm to 800 nm. Molar extinction coefficients at 280 nm were calculated by normalizing the absorbance value at 280 nm to DOC.In addition,we also included values for the specific UV absorbance at 254 nm,which is typical for the aromatic groups with varying degrees of activation(Korshin et al., 2009). SUVA254(UV-adsorption A254nm(1cm)/mg C L-1) is defined as the UV absorbance of a given sample at 254 nm divided by the DOC concentration of the sample. A253/A203is the ratio of UV-Vis absorbance at 253 nm to that at 203 nm. The A253/A203index can reflect the number of substituent in a compound,and higher A253/A203ratios correspond to higher concentrations of substitution groups(Li et al.,2014).The spectral slope ratio(SR)is the ratio of the spectral slope of short wavelength (275-295 nm)to that of long wavelength (350-400 nm), and its variation is related to different molecular weights of DOM (Zhang et al.,2016;Helms et al.,2009).

2.3.3 Fluorescence

Filtered samples were diluted with Milli-Q water to achieve absorbance values<0.05 at 254 nm. This dilution avoids the need to do additional innerfilter corrections. All EEMs were measured using a Hitachi F-7000 fluorescence spectrophotometer (Hitachi, Japan) equipped with a 700-V Xenon lamp as an excitation source.Fluorescence EEM spectroscopy involved scanning and recording individual emission spectra(200-600 nm)at sequential 10 nm increments of excitation wavelength between 250 nm and 400 nm as described by Huguet et al(2009). Fluorescence regional integration(FRI)technique which integrates fluorescence intensity values in five divided excitation-emission regions (Chen et al., 2003). In the present study, peaks at shorter excitation wavelengths (<250 nm)and shorter emission wavelengths (<380 nm)are related to simple aromatic proteins, such as tyrosine and tryptophan-like compounds (regions I and II). Peaks at shorter excitation wavelengths (<250 nm)and longer emission wavelengths(>380 nm)are related to fulvic acid-like substances(region III).Peaks at intermediate excitation wavelengths (250-280 nm)and shorter emission wavelength (<380 nm)are associated with soluble microbial byproduct material (region IV).Peaks at longer excitation wavelengths (>250 nm)and longer emission wavelengths (>380 nm) represent humic acid-like organic fractions (region V). Fluorescence spectra can be used to detect the unidentifiable DON since some of it is incorporated into humic substances(Zhang et al.,2016; Pehlivanoglu et al.,2006). To be noted,FRI technique is semi-quantitative. That is to say,the results derived from the FRI technique represented relevant proportion of each component, instead of their absolute content.

2.4 Data analysis

Each sample was measured in triplicate,and our results are based on the arithmetic mean. Data were processed using Excel 2010, and SPSS 19.0 was used for correlation analysis. FRI was analyzed using Matlab 2015a(Zhang et al.,2015). The regional distribution of pore water nutrients was drawn using Surfer12.

3 Results and discussion

3.1 Characteristics of pore water DON in Erhai Lake sediment

3.1.1 Characteristics of pore water DON in Erhai Lake sediment

The DON concentration of sediment pore water samples varied greatly among sampled regions in Erhai Lake,ranging from 0.198 mg/L to 1.710 mg/L, with an average of ca. 0.773mg/L. The DON distribution in sediment pore water samples followed the pattern of middle(1.449mg/L)>south(0.828 mg/L)>north(0.266 mg/L)(Fig.2a). Sediment pore water DON concentration was nearly four times higher than the DON concentration in the overlying water (mean value of 0.20 mg/L).The steep gradient at the sediment-water interface in Erhai Lake and the salt wedge pump may have led to the input of endogenous pollutants from the sediment to the pore water (Isaac et al., 2012). The distribution, components, and sources of sediment pore water DON are largely influenced by biological and non-biological factors,including meteorological,geological,and hydrological conditions (Yang et al.,2016). However,pore water nitrogen concentration may not be related to sediment contaminant status since the highest pore water DON concentrations were found in the least polluted areas.

The spatial distribution of DON in sur?cial sediment was site-speci?c,following the pattern of south(44.89 mg/kg)>north(35.24 mg/kg)>middle(32.08 mg/kg)(Zhang et al.,2016). This shows that there are environmental factors that affect sediment DON movement to pore water in the middle region. The middle region is the deepest(with a mean depth of 20.5m),and the pressure was likely correspondingly the highest in this area. This may have accelerated DON mineralization from sediment to pore water. Alternatively,the central sediments had the distinct properties from other regions of Erhai Lake,since the sediment deposition rate in the central region was clearly different from northern and southern area (Ni et al., 2016). Consequently, the sediment properties were different,further leading to the regularities of distribution being inconsistent among the whole lake.

Fig.2 Spatial distribution of DON in Erhai Lake’s(a)pore water,(b)overlying water,and(c)sediment.

3.1.2 Vertical distribution of DON in Erhai Lake sediment pore water

The vertical variation of sediment pore water DON is shown in Figure 2. For the convenience of analysis, the pore water columns were divided into three layers: a surficial layer of 0-8 cm,an interim layer of 8-20 cm,and a bottom layer of 20-30 cm. The sediment pore water DON content profile was affected by location (Fig.2),ranging from 0.46 to 5.67mg/L in the northern part of the Lake, with an average concentration of 3.06mg/L.Sediment pore water DON exhibited a declining trend with depth. The pore water DON concentration gradually decreased from the surface to the bottom for the northern and southern samples. The opposite trend occurred in the middle region; with the increase of pore water depth, DON had a gradual increasing trend. Moreover, the underlying DON content was nearly eight times higher than the surficial pore water,which is similar to the pore water results found by Burdige et al. (2001).

Pore water DON is likely to be affected by Erhai Lake’s unique geographical environment. The sediment DON content are controlled by a balance between DON production and consumption (Burdige et al., 2001).Consequently, in recent years (after the 1950s), the human activity accelerated the biodegradation and transformation of the sediment pore water DON in the central area of Erhai Lake,showing higher concentrations at deeper layers.

3.2 Compositional characteristics and sources of pore water DON

3.2.1 Structural characteristics of pore water DON

The A253/A203ratio was very low in pore water DON of Erhai sediment (Table 1). This suggests that the sediment pore water DON was dominated by chains, with a small number of aromatic substituents (Korshin et al.,1997). In Erhai Lake sediment pore water,the SUVA254value ranged from 0.20 to 0.59(mean value of 0.41),which was lower than the SUVA254value from Wang et al. (2013)(mean value of 0.58). This indicates that the Erhai Lake sediment pore water DON had few conjugated double bonds and relatively low aromaticity (Zhang et al.,2015). However,the SUVA254value was less than 3,suggesting the major pore water DON components were hydrophilic(Anu et al.,2011).

Fig.3 Change in DON content in pore water of Erhai Lake sediment with time.

Helms et al. (2009) reported SR>1 for microbial-derived samples, and this was inversely related to the molecular weight (MW) of the DOM in a water sample. In this study, SRvalues were all above 1 (Table 1),which followed the trend of northern (1.97)>middle(1.71)>southern(1.41), suggesting the pore water DON in Erhai sediment was generally from microbial sources,and their MW decreased progressively from north to south of Erhai Lake. It is possible that the adsorption or precipitation of high MW DOC has occurred along the water flow direction due to humification (Lepane et al., 2004). The spatial distribution of MW was consistent with the direction of water flow in Erhai Lake,demonstrating that the upstream region of the Lake was likely to have high MW sediment pore water DON.

FRI can reveal the configuration and heterogeneity of DOM,and it has been widely used to quantitatively analyze wavelength dependent fluorescence intensity data from EEM spectra(He et al.,2011). The distributions of Pi,nfor the five EEM regions in DON and its corresponding fractions isolated from porewater of different location and depth are presented in Table 1. Humus substances (PIII+V, 53%-78%) accounted for the majority of DON; protein-like components (PI+II, 12%-23%) were similar to microbial products (PIV, 13%-19%).The Pi,nvalues for Regions I and IV in the pore water DON decreased rapidly then stabilized with depth increment. ThePi,nvalues for Regions III and V significantly increased as the aromatic protein-like compounds were depleted (Fig. 4). The above information illustrated that the partial protein components which were easy to be degradation in sediment pore water DON were declined fast, but humus-like substances were persistent,then further stored in the deeper sediment pore water,which was refractory to degraded by microorganisms. In the overlying water DON,the proportion of protein-like components (17%)and microbial products (22%)was higher, and the proportion of humus substances (61%) was lower than the components of the sediment pore water DON.This result may reflect the degradation processes of DON in overlying water and the settling of the humus form of DON in the sediment pore water.

3.2.2 Source characteristics of pore water DON in lake sediment

Fluorescence index(FI)values can be used to distinguish DON sources. The FI terrigenous and endogenous endpoint values are 1.4 and 1.9,respectively(Li et al.,2014). The FI mean value in Erhai Lake sediment pore water was 1.84, suggesting the sediment pore water DON was mainly derived from aquatic and microbial sources(Huguet et al., 2009). The biological index (BIX) ranged from 0.88 to 0.99, suggesting sediment pore water DON was autochthonous and came mainly from degraded organic matter in sediments and degraded submerged plants and plankton,corresponding with the FI results reported by Huguet et al. (2009).

The content and activity of microbes are related with DON degradation (Zhang et al., 2014), which may affect the release of sediment DON. Microbial activity also affected the concentration of pore water DON. Anumber of studies have shown that microorganisms are closely related to N metabolism. For example, Bowles et al. (2014) suggested that fungal communities might increase agroecosystem N retention. Lei et al. (2012)suggested that actinomycetes played a vital role in N cycling. In Erhai Lake,sediment pore water DON content was positively correlated with sediment bacteria, actinomycetes, fungi, and microbial N biomass, with correlation coefficients of 0.67, 0.81, 0.83, and 0.70, respectively (Table 2). This indicated that increased microbial activity would increase the content of sediment pore water DON, and the control of microbial activity can be manipulated to enhance nitrogen nutrient cycling capacity,which may help to adjust and control sediment pore water DON.Zhang et al. (2015)found that actinomycetes was the dominant microorganism in Erhai Lake sediment,and the DON content in Erhai Lake was endogenous, revealing that the forms of DON were affected by actinomycetes activity.

Table 1 UV-vis spectra parameters and percentage distribution(fi×100%)of pore water DON in Erhai Lake sediment.

We also evaluated sediment enzyme activity involved in N assimilation to investigate the preferred DON utilization capabilities in sediment. The correlation analysis results showed that sediment protease,peroxidase,and urease activity were significantly and negatively correlated with sediment pore water DON,with correlation coefficients of-0.54,-0.71,and-0.64,respectively. In addition,DON was significantly and positively correlated with polyphenoloxidase and invertase activity (Table 2). These results indicate that sediment enzyme activity was closely correlated with the nitrogen level in the lake sediment pore water. Polyphenoloxidase and invertase increased the DON content in sediment pore water, and the protease, peroxidase, and urease accelerated the depletion of sediment pore water DON.

Sediment pore water DON content showed no clear correlation with the f450/f500index(FI).It was significantly and positively correlated with BIX and HIX in the middle samples,with correlation coefficients of 0.928(p<0.01)and 0.818(p<0.05),respectively. Huguet et al. (2009)found that the HIX index was positively correlated with the aromaticity of DON,the BIX index corresponded to the characteristic of autochthonous biological activity,and FI could discriminate DOM sources. This suggested that sediment pore water DON(which had the highest concentration in the middle samples) was affected by autochthonous biological activity and possessed high aromaticity. In contrast, the sediment pore water DON concentration was negatively correlated with BIX in the south(p<0.05)and positively correlated with HIX in the north(p<0.05). This implies that the source ofsediment pore water DON in the south and north was different from that of the middle region. Sediment pore water DON in the south was more likely affected by terrigenous sources, whereas the north had more complex molecules like high molecular weight aromatics(Huguet et al.,2009).

Table 2 Correlation of pore water DON with microorganisms and enzymes in Erhai Lake sediment.

3.3 Changes in pore water DON in Erhai Lake sediment

Eutrophication caused by excessive nutrient discharge is a global challenge. The ecological implications of human alterations to the nitrogen cycle have been profound as human activities have already significantly altered the global nitrogen cycle(Zhang et al.,2014). Many studies have indicated that non-point source pollution has played an increasingly significant role in water quality deterioration in China(Li et al.,2014). Previous research has found that agricultural input was the main nitrogen source in Erhai Lake(Ni et al., 2015). Erhai Lake has recently undergone a transformation from mesotrophic to eutrophic conditions due to the wide use of chemical fertilizers and the destruction of wetland vegetation along the lakeshore (Zhang et al., 2015). The problem of growing sufficient food has been largely solved by agricultural science, and the application of nitrogen as fertilizer in the Erhai Lake Basin has been a substantial contributor to increased productivity. From 1997 to the early 2010s, anthropogenic creation of reactive N compounds increased globally from approximately 59.5 to 90Tg N. The amount of nitrogen applied was more than three times crop productivity. A large amount of organic nitrogenous fertilizer likely flowed to the Lake,resulting in an increase in sediment pore water DON(Ni et al.,2016).

An average sedimentation rate of 0.24 mm/year in the south and north was calculated from radiocarbon calculations(14C andδ13C)(Zhang et al.,1993). The sedimentation rate was 0.16 mm/year in the middle region(Zhang et al., 1993). The relationship between DON components and sample age of the three cores collected from different areas was analyzed to reconstruct the historical record of pore water DON and to elucidate the effect of natural and anthropogenic processes on different areas of Erhai Lake in the past decades(Fig. 4). The temporal distribution of DON can be divided into two periods that most closely relate to the history of local natural and anthropogenic activities. Previous research has also divided the changes in Erhai Lake into these two time periods based on pore water DON vertical distribution (Zhang et al., 2015). Before the 1970s, the humic ratio was stable and showed no obvious trend, with a mean proportion of 72%, 78%, and 75% in the north,south,and middle,respectively. From the 1970s to the 2010s,DON content showed a dramatic decrease(55%,61%,and 67%,respectively)(Fig.4b). Protein-like and microbial by product-like compounds showed the opposite trend(Fig.4c). This indicates that DON compounds could record the changes in Erhai Lake over time,since the humic-like component was refractory to degradation(Xu and Zheng,2003).

This study also explored changes in the fluorescence index with sample age; the results showed that the HIX trend was similar to the PIII+Vof FRI trend (Fig.4a,b). However, BIX showed a different response. The mean value of BIX was approximately 1 before the 1970s,and it increased slightly after the 1970s. This implies that the sediment pore water DON was predominantly of autochthonous origin (Huguet et al., 2009), and the proportion of autochthonous DON has increased over the past 40 years.

The characteristics of DON components corresponded to human activities. The process of eutrophicationin Erhai Lake accelerated from the beginning of the 1970s. Before the 1970s, the Erhai Lake Basin was under development. During this time,when there was only a small effect of human activities on Lake Erhai, the nutrient content of the sediment was stable,and the Lake’s nutritional status was dominated by natural circulation.Submerged macrophyte communities rapidly spreaded throughout the Lake after the operation of the Xi’erhe hydropower station in 1972 (Pan et al., 1999; Ni et al., 2016). After the 1990s, Erhai Lake once again experienced considerable changes,with submerged macrophyte communities suffering from large scale deterioration,leading to nutrients dramatically accumulating in Lake sediment (Wang et al., 2015). Consequently, the labile and high bioavailable components of sediment pore water DON,such as protein-like and microbial byproductlike compounds, increased dramatically, enhancing the eutrophication risk for Erhai Lake. The pollution was

endogenous in the Erhai Lake Basin, which was under the influence of microorganisms and enzyme activity affected by sediment substrates and N characteristics (Zaveri et al.,2016;Giagnoni et al. 2016;Seifert-Monson et al.,2014).

4 Conclusions

The DON distribution in sediment pore water of Erhai Lake followed the pattern of middle(1.449 mg/L)>south(0.828 mg/L)>north(0.266 mg/L).Humic substances(PIII+V,53%-78%)accounted for the majority of the pore water DON; protein-like components (PI+II, 12%-23%) were similar to microbial products (PIV, 13%-19%).The sediment pore water DON came mainly from the degradation of organic matter released from sediment,as well as the degradation of submerged plants and plankton. The DON compounds recorded changes in Erhai Lake over time. The humic-like component was refractory to degradation,and it could reconstruct the historical record of pore water DON.

To improve Erhai Lake’s aquatic ecosystem, the nitrogen input to the Lake must be managed. Our study suggests that sediment pore water has the potential to release more DON in Erhai Lake;future research should further investigate this pollution risk.

Acknowledgements

This research was supported by the Beijing Major Science and Technology Projects(Z181100005318001), Beijing Natural Science Foundation(8192004),the National Natural Science Foundation of China(41503113).