Introduction

Shallow lakes are particularly susceptible to becoming contaminated by organic waste produced through human activities. Adding organic matter to a lake can increase the phytoplankton biomass and increase primary production, causing eutrophication. It is therefore necessary to understand the types and origins of organic matter present in shallow lakes.

Sources of various components of organic matter (e.g., cellulose, lignin, proteins, hydrocarbons, and humus) in aquatic systems have been described previously (Brown et al. 1972; Hedges et al. 1988; Saliot et al. 1988; Fogel et al. 1989; Fichez Mf et al. 1993; Ribeiro et al. 2011; Pradhan et al. 2014). The oxidation products of lignin and stable carbon isotopes of local plants and suspended particulate material were analyzed to study the sources of organic matter (Hedges et al. 1988). Sterols and alcohols of SOM were measured to evaluate its origins (Ribeiro et al. 2011). Lignin phenol, C/N elemental concentration ratio, and δ13C of SOM were determined to investigate the sources of organic matter on the west coast of India (Pradhan et al. 2014)

Sources of organic matter can be identified by analyzing stable isotopes in the organic matter. The stable carbon isotope ratios (δ13C) and stable nitrogen isotope ratios (δ15N) in aquatic organisms can serve as useful indicators of anthropogenic activities (Cremona et al. 2009; Karube et al. 2010; Hoffman et al. 2012). δ13C and δ15N ranges for different sources overlap somewhat, and various anthropogenic sources of nitrogen can affect the δ15N of aquatic organisms, and thus δ13C and δ15N data cannot be used except in combination with other tracers (Kendall et al. 2001). Stable isotopes (i.e., δ13C and δ15N) have frequently been used as proxies to identify environmental changes in lakes (Brenner et al. 1999; Herczeg et al. 2001; Wu et al. 2006; Torres et al. 2012). The stable isotopes of sulfur, nitrogen, and carbon were used to trace organic matter flow in salt marshes and estuarine waters (Peterson and Howarth 1987). The distribution and compound-specific carbon isotope ratios of n-alkanes and fatty acids were examined to investigate organic matter sources in a eutrophic lake (Fang et al. 2014).

Organic matter in lake sediment forms part of the historical record preserved in the sediment. Lakes with different species and watersheds have SOM with distinct biogeochemical differences (Meyers and Ishiwatari 1993; Brenner et al. 1999). Authigenic organic matter in lakes and terrestrial detritus surrounding a lake are the main sources of organic matter found in lake sediment. Different components of lake SOM retain information on the sources of the SOM. The isotopic composition of SOM reflects the isotopic compositions of the sources of the SOM. δ13C and δ15N signatures have been used to distinguish between different types of organic matter. For example, δ13C and δ15N of SOM were used as tracers for changes in trophic states in lakes in Florida, USA (Brenner et al. 1999). δ13C and δ15N of suspended particulate matter and sediment from near-shore areas of the Tanzanian part of Lake Victoria were determined to identify the various sources of organic matter found in different parts of the lake (Machiwa 2010).

Poyang Lake, which is the largest freshwater lake in China, is a shallow and seasonal lake. It is in the northern part of Jiangxi Province, in the middle and lower reaches of the Yangtze River. Poyang Lake is influenced by the Yangtze River and five other major rivers (the Ganjiang River, Fuhe River, Xinjiang River, Raohe River, and Xiushui River). Poyang Lake has alternating wet and dry periods. A river appears in the middle of Poyang Lake during the dry period, and some relatively closed shallow lakes appear along the edge of Poyang Lake. The main part of Poyang Lake is generally not strongly affected by pollutants, but water exchange in the shallow lakes is weak because of sluggish water flow, so these lakes are more strongly influenced by human activities and nutrient oversupply. However, little research has been performed on these closed shallow lakes. It is therefore very important that these lakes are studied.

In this study, we used the C/N ratio and the δ13C and δ15N values for surface SOM from the lakes to identify the sources of organic matter and nitrogen to the small closed shallow lakes alongside Poyang Lake. We studied the shallow lakes Bang Lake, Dahuchi Lake, Sha Lake, and Zhu Lake. Bang Lake (7300 hm2), Dahuchi Lake (3000 hm2), and Sha Lake (1400 hm2) are in the National Poyang Lake Nature Reserve, which contains nine small lakes and the surrounding meadows in the northwestern part of Poyang Lake. The three lakes are separated from Poyang Lake by a natural dam when the lake retreats in the dry season, and the average depth of the lakes is 0.5–1.0 m. Zhu Lake, which has an average depth of 5–6 m, is to the east of Poyang Lake. Zhu Lake was previously part of Poyang Lake but is now a reservoir-type lake that was cut off from Poyang Lake by a dam that was built in the 1960s.

Zhu Lake was also studied to allow us to gain a more complete understanding of the sources of organic matter in the area by comparing the results for Zhu Lake with the results for the lakes in the National Nature Reserve. Two urban lakes (Qingshan Lake and Xiang Lake) were also studied and the results compared with the results for the shallow lakes alongside Poyang Lake.

Methods

Sample Collection

Surface sediment samples (0–10 cm deep) were collected from the small shallow lakes Bang Lake, Dahuchichi Lake, Sha Lake, and Zhu Lake in July 2015. The sampling sites are shown in Fig. 1. To allow results for different types of lakes to be compared, we also collected surface sediment samples from the two urban lakes Qingshan Lake and Xiang Lake, in Nanchang. Bang Lake, Dahuchi Lake, and Sha Lake (in the northwestern part of Poyang Lake) are within the Jiangxi Poyang Lake National Nature Reserve. Zhu Lake, which is east of Poyang Lake, is a reservoir-type lake connected to Poyang Lake through a sluice. Fish are farmed in Zhu Lake. Qingshan Lake and Xiang Lake are urban lakes in Nanchang. Qingshan Lake is in the downtown area and is surrounded by cement slopes. Xiang Lake is much larger than Qinshan Lake and is surrounded by aquatic macrophytes, such as reeds and lotus. The samples were immediately transported to the laboratory. Each sample was then freeze-dried in a vacuum freeze dryer, ground, passed through a 200-mesh sieve, and sealed in a polythene bag.

Fig. 1
figure 1

Map of Poyang Lake showing the locations of the sampling sites

Full size image

Elemental Analysis and Stable Isotope Analysis

The total organic carbon (TOC) content of each sediment sample was determined after treating a sample of the sediment with 2 M KCl to remove carbonates. The total organic nitrogen (TON) content was determined after treating a sample with 0.1 M HCl to remove inorganic nitrogen. The TOC and TON contents were determined using an elemental analyzer (2400IICHNS/O, Perkin-Elmer, Foster City, CA, USA), which had an analytical precision of 0.1%. The C/N ratio for organic matter was calculated from the TOC and TON data. δ13C and δ15N were determined using a gas isotope ratio mass spectrometer (Finnigan MAT 252, Thermo Fisher Scientific, Waltham, MA, USA) after a sample had been purified using liquid nitrogen.

Results

The TOC and TON contents of the surface sediment samples from the lakes around Poyang Lake were 0.296%–1.877% and 0.065%–0.185%, respectively, as shown in Table 1. The samples from Bang Lake, Sha Lake, Zhu Lake, and Dahuchi Lake had the highest TOC contents, the average TOC contents being 1.14% ± 0.34% (n = 19), 0.78% ± 0.21% (n = 9), 0.77% ± 0.23% (n = 10), and 0.66% ± 0.14% (n = 9), respectively. The average TON contents of the Bang Lake, Sha Lake, Zhu Lake, and Dahuchi Lake samples were 0.14% ± 0.03%, 0.12% ± 0.02%, 0.12% ± 0.03%, and 0.11% ± 0.02%, respectively. The average TOC contents of the Qingshan Lake and Xiang Lake samples were 1.86% ± 1.79% (n = 6) and 2.33% ± 0.61% (n = 5), respectively. The average TON contents of the Qingshan Lake and Xiang Lake samples were 0.16% ± 0.14% and 0.19% ± 0.08%, respectively. It is clear from these data that Bang Lake is more seriously contaminated with organic pollutants than the other lakes alongside Poyang Lake and that the urban lakes are even more polluted. It is important to gain an understanding of the types and origins of the organic contaminants of the lakes. We measured the C/N ratio and the δ13C and δ15N values to achieve this.

Table 1 TOC, TON, C/N, δ13C, and δ15N of the surface sediment samples
Full size table

The C/N ratio has often been used to indicate the primary sources of SOM. Terrestrial plants have high lignin and fiber contents, unlike protein-rich phytoplankton. Algae usually have C/N ratios of 4–10, but vascular plants have C/N ratios of 12 or more. Sewage, an anthropogenic source of C and N to lake sediment, has C/N ratios of 6.6–12.6. The C/N ratios in the surface sediment samples from the lakes alongside Poyang Lake were 4.54–10.17. The average C/N ratio for Bang Lake (7.98 ± 1.00) was a little higher than the average C/N ratios for Sha, Zhu, and Dahuchi Lake. The average C/N ratios for Qingshan Lake and Xiang Lake were 12.07 ± 9.52 and 13.53 ± 3.50, respectively. These data suggest that the SOM in Bang Lake is affected more by external factors than SOM in the other lakes alongside Poyang Lake and that Qingshan Lake and Xiang Lake are more influenced by anthropogenic sources than the other lakes.

The differences between the δ13C values for the SOM of most lakes, caused by different carbon sources, were minor. The δ13C values for the lakes alongside Poyang Lake were between −29.66 and −22.18‰, and the Bang Lake SOM was a little depleted in 13C. The average δ13C values for Qingshan Lake and Xiang Lake were −23.40‰ ± 3.43‰ and −26.26‰ ± 0.33‰, respectively. It can be seen from Table 1 that the urban lakes, especially Qingshan Lake, were more enriched in 15N than were the lakes alongside Poyang Lake. Of the lakes alongside Poyang Lake, Bang Lake and Zhu Lake were a little more enriched in 15N than were Sha Lake and Dahuchi Lake.

Discussion

Assessment of the Nutrients in Surface Sediment

Changes in the TOC and TON contents of authigenic organic matter in a lake can indicate changes in primary productivity and biomass. The TOC and TON contents of the surface sediment of a lake reflect the types and intensities of human activities affecting the lake.

The organic index (= TOC (%) × TON (%)) and organic nitrogen content (Table 2) can be used as proxies to assess the nutrient level of a surface sediment (Neufeld and Hermann 1975; Lin et al. 2009; Gan et al. 2012; Yang et al. 2014; Meng et al. 2015). The organic index is commonly used to assess the environmental conditions of sediment, and the organic nitrogen content is often used to determine the degree to which nitrogen pollution affects the surface sediment of a lake. The organic indices and organic nitrogen contents of the sediment samples from our sampling sites are shown in Tables 2 and 3.

Table 2 Evaluation criteria for the organic index and organic nitrogen content
Full size table
Table 3 Organic indices and organic nitrogen contents of the sediment samples
Full size table

Using the organic indices to evaluate the eutrophic statuses of the lakes, we found that Poyang Lake and the urban lakes are relatively clean and that the urban lakes had higher organic indices. This agreed with the results of a previous study (Lin et al. 2009) in which Poyang Lake was found to be relatively free of pollution.

Using the organic nitrogen contents, we found that Sha Lake, Dahuchi Lake, and Zhu Lake are relatively clean and that Bang Lake and the urban lakes are contaminated. This indicates that Bang Lake and the urban lakes are more polluted with nitrogen than the other lakes. The nitrogen in the urban lakes was derived from urban sewage, and the nitrogen in the lakes alongside Poyang Lake was supplied by the surrounding agricultural area.

Bang Lake, Sha Lake, and Dahuchi Lake are adjacent to each other in Jiangxi Poyang Lake National Nature Reserve. We found that both proxies indicated that Bang Lake is more polluted than Sha Lake and Dahuchi Lake. Sha Lake and Dahuchi Lake are closely surrounded by floodplain levees, low hills, and natural sand dykes. In contrast, Bang Lake does not have strongly defined edges. During high water periods, Bang Lake becomes connected to Ya Lake, which is to the southwest and next to Gongqing City. This means that Bang Lake is more affected than Sha Lake and Dahuchi Lake by human activities and explains why the surface sediment in Bang Lake is more contaminated than the surface sediment in Sha Lake and Dahuchi Lake. Unlike Bang Lake, Sha Lake, and Dahuchi Lake, Zhu Lake is not in the Jiangxi Poyang Lake National Nature Reserve. Zhu Lake is a relatively closed lake that is used to farm fish and is connected to Poyang Lake through a sluice. The organic index was higher for Zhu Lake than for Sha Lake and Dahuchi Lake but lower than for Bang Lake. Qingshan Lake and Xiang Lake are in Nanchang. The organic index and organic nitrogen content were higher for Xiang Lake than for Qingshan Lake.

Sources of Organic Matter to the Surface Sediment

The TOC and TON contents varied in similar ways, as shown in Fig. 2, indicating that the organic matter in the surface sediment in the different lakes had similar sources. There was an exception in that the TOC content was high but the TON content low in the sample from site Q3 (Qingshan Lake). There is a sewage outfall near site Q3, and the wastewater released here adds anthropogenic organic matter to Qingshan Lake, altering the characteristics of the organic matter nearby.

Fig. 2
figure 2

TOC and TON in the surface sediment samples. (a: Bang Lake; b: Sha Lake; c: Dahuchi Lake; d: Zhu Lake; e: Qingshan Lake; f: Xiang Lake)

Full size image

The organic matter sources are interpreted as a mixture of: (1) phytoplankton, (2) aquatic macrophytes, (3) terrestrial plants detritus, (4) soil organic matter, (5) anthropogenic inputs. The sources of organic matter and nitrogen to the surface sediment samples from the lakes were closely related to disturbances caused by human activities. When human influences are weak, the organic matter and nitrogen will mostly come from authigenic sources, such as phytoplankton and aquatic macrophytes. In contrast, when human influences are strong, the organic matter and nitrogen in the sediment samples are affected by both autotrophs and allochthonous material.

The C/N ratio reflects the original proportion of autotrophs and allochthonous material in a sample. The C/N ratios for phytoplankton, benthic algae, and aquatic macrophytes are about 5–8, about 8–10, and about 10–30, respectively (Thorp et al. 1998; Kendall et al. 2001). The C/N ratios for terrestrial plants and terrestrial arable soils are about 15–30 and about 10–12, respectively. The C/N ratios for terrestrial and aquatic plants overlap considerably. Sewage, an anthropogenic source, has C/N ratios of 6.6–12.6 (Machiwa 2010). The isotopic composition of carbon in SOM reflects the productivity of a lake and the sources of the organic matter. The average δ13C values for C3 plants, C4 plants, phytoplankton, and macrophytes are about −27‰, about −14‰, between −32 and −23‰, and between −27 and −20‰ (Kendall et al. 2007). The δ15N values for most terrestrial plants are in the range − 6‰ to +5‰ (Fry 1991), the δ15N values for most soil organic matter are in the range + 2‰ to +5‰ (Broadbent et al. 1980), and the δ15N values for macrophytes and algae vary widely, between −15 and +20‰ and between +5 and +8‰, respectively (Boutton 1991; Hamilton and Lewis 1992; Angradi 1993; Angradi 1994; Thorp et al. 1998; Cloern et al. 2002; Kendall et al. 2007). Sewage has δ15N values of +7‰ to +25‰, δ13C values of −26.7‰ to −22.9‰, and C/N ratios of 6.6 to 12.6 (Machiwa 2010). The main types of organic matter in lake sediment have very variable δ13C and δ15N values and C/N ratios (Table 4).

Table 4 Typical compositions of the main sources of sedimentary organic matter
Full size table

The δ13C and δ15N values for different sources overlap somewhat and can be influenced by various anthropogenic sources of nitrogen. C/N ratios are therefore normally more helpful than δ13C and δ15N values for identifying the dominant source of organic matter (Kendall et al. 2001).

We used the δ13C and δ15N data and the C/N ratios to investigate the sources of organic matter to the lakes we studied. As is shown in Fig. 3a, the SOM in Sha Lake and Dahuchi Lake clearly originated in phytoplankton. The characteristics of the SOM in Bang Lake were different from the characteristics of the SOM in Sha Lake and Dahuchi Lake, and the SOM in Bang Lake was influenced to a significant degree by soil organic matter. As mentioned above, Bang Lake does not have a well-defined edge, and thus soil surrounding Bang Lake becomes washed into the lake. The SOM in Bang Lake was therefore affected by both phytoplankton and soil organic matter. The δ13C values of SOM increase as the trophic state increases (Machiwa 2010). The δ13C values for the Bang Lake sediment, which had a relatively high organic nitrogen content, were lower than the δ13C values for the Sha Lake and Dahuchi Lake sediment. This was because SOM in Bang Lake is affected by soil organic matter. Terrestrial plant detritus washed into the lake with soil will have decreased the δ13C value for the SOM in Bang Lake. The SOM in Zhu Lake was also found to be affected by both phytoplankton and soil organic matter. However, the δ13C values were higher for the Zhu Lake SOM than for the Bang Lake SOM, indicating that Zhu Lake has a higher level of primary production than does Bang Lake. Phytoplankton prefer to photosynthesize using aqueous CO2, which is richer in lighter isotope 12C, rather than carbonates, which are richer in the heavier isotope 13C. During periods of high primary productivity, CO2 in the photic zone may be exhausted, forcing phytoplankton to use bicarbonate, which is enriched in 13C. The SOM in a lake with a high degree of productivity will therefore be enriched in 13C. Fish feed added to Zhu Lake because of fish farming provides nutrients for phytoplankton and has made Zhu Lake more productive than Bang Lake, Sha Lake, and Dahuchi Lake. The SOM in Xiang Lake, in Nanchang, originated in aquatic macrophytes, as shown in Fig. 3a, matching our observations that there were many reeds and lotuses around the sampling sites in Xiang Lake. It was clear that the SOM in Qingshan Lake, which is in the center of Nanchang and surrounded by cement slopes, was affected by sewage.

Fig. 3
figure 3

C/N ratios and δ13C and δ15N values for the surface sediment samples

Full size image

Nitrogen in sediment in shallow lakes is mainly present as organic nitrogen (Wu et al. 1996; Kendall et al. 2007; Meng et al. 2015), and primarily originates in limnological organisms under natural conditions. Fertilizer application, wastewater discharge, and the mineralization, nitrification, and denitrification of organic nitrogen, amongst other processes, can readily affect the dissolved inorganic nitrogen in sediment. The δ15N values of sediment have been found to be largely determined by the isotope ratios in the original materials (Kumar et al. 2004; Xiao and Liu 2004). We therefore assessed the sources of nitrogen in SOM by studying the organic nitrogen. The isotopic compositions of aquatic organisms is determined by the isotopic composition of dissolved inorganic nitrogen, and the δ15N values of aquatic organisms strongly influence the δ15N value of SOM. Variations in the nitrogen isotopic compositions of SOM can therefore indicate the degree to which human influences have disturbed the SOM composition.

Different conclusions were drawn about the sources of SOM (e.g., the sources of SOM to Bang Lake, Sha Lake, and Dahuchi Lake being soil organic matter or terrestrial plants) from Fig. 3b than were drawn from Fig. 3a. This indicates that nitrogen isotopes cannot be used effectively to determine the sources of organic matter. However, we could estimate the impacts of anthropogenic sources using the nitrogen isotope measurements. The δ15N values for Sha Lake and Dahuchi Lake were lower than the δ15N values for Bang Lake and Zhu Lake, indicating that Sha Lake and Dahuchi Lake are less affected by human activities than Bang Lake and Zhu Lake. These conclusions agreed with the conclusions drawn earlier. The δ15N values for Qingshan Lake were relatively high (+4.6‰ to +8.1‰), suggesting that SOM in this lake is strongly influenced by municipal wastewater. The SOM from Xiang Lake had a relatively high organic nitrogen content, as shown in Table 1, indicating that Xiang Lake is more eutrophic than the other lakes. In general, the δ15N value for plankton increases from oligotrophic to eutrophic lakes (Brenner et al. 1999). However, the δ15N values for Xiang Lake (+3.76‰ to +5.21‰) were lower than the δ15N values for Qingshan Lake. This can be explained by the greater relative contribution of 15N depleted, nitrogen-fixing cyanobacteria to the sediment organic matter in the hypereutrophic Xiang Lake having a tendency toward dominance by cyanobacteria.

Conclusions

The results of the study suggest that the TOC and TON contents, C/N ratios, and stable isotope signatures of SOM could be used as proxies for the trophic state of a lake and the sources of organic matter to the lake. The higher the TOC and TON contents are, the more eutrophic a lake is. The δ15N value for SOM increases as the effects of human activities increase, and the δ13C value increases as the primary productivity of a lake increases. Despite the large overlap in the C/N ratio, δ13C, and δ15N datasets, it was possible to identify the sources of organic matter to the surface sediment samples in the lakes we studied.

Sha Lake and Dahuchi Lake, in the National Poyang Lake Nature Reserve, were found to be relatively clean, and the low δ15N values for the SOM indicated that these two lakes are affected less than Bang Lake and Zhu Lake by human activities. The SOM in these two lakes clearly originated in phytoplankton. Bang Lake, also in the reserve, is more affected by human activities, and Zhu Lake (east of Poyang Lake) is affected by anthropogenic activity because fish farming is performed in this lake. The SOM from Bang Lake and Zhu Lake originated in both phytoplankton and soil organic matter. The δ13C values were higher for Zhu Lake than for Bang Lake, indicating that Zhu Lake has a higher degree of primary productivity. The δ15N values were slightly higher for Bang Lake and Zhu Lake than for Sha Lake and Dahuchi Lake, indicating that these two lakes are influenced by human activities.

Qingshan Lake and Xiang Lake, both in Nanchang City, are eutrophic. The δ15N values for Qingshan Lake suggested that the SOM is strongly affected by sewage organic matter. The δ15N value for plankton was lower in Xiang Lake than in Qingshan Lake because of N2 fixation by cyanobacteria in hypereutrophic Xiang Lake.