Phosphorus (P) forms, with respect to sediment characteristics, and an in-stream sluice were studied in the river–lake system, Huai River catchment area, China. The mean of total P in sediments in the mainstream of the Huai River was higher than that in the Hongze Lake. It was found that P fractions varied in the sediments throughout the river and lake. Detrital-P was the dominant P fraction in the mainstream and organic P and detrital-P were the dominant P fractions in the lake, which could indicate: biologically available and non-biologically available forms. Useful information for the interpretation of P fractions could also be obtained from major sediment characteristics. Whether the relations between P fractions and grain size characteristics were significant or weak, 0.125 mm was a threshold grain size for P fraction distribution in sediment. In addition, the Bengbu Sluice, one of the most important in-stream facilities in the Huai River catchment area, regulated not only the transport of P in sediments upstream and downstream of the sluice, but also the distribution of P fractions in the river–lake system. Therefore, it was confirmed that nutrient loadings could be prevented from reaching the watershed, as well as improved ecological diversity by integrating sluice operation.

INTRODUCTION

The river–lake system can be considered as the longitudinal composition of two different areas where lakes are hydrologically linked with stream or river courses in the sense that the lake is fed by river waters and vice versa (Hillbricht-Ilkowska 1999). The river has the most important function as collection, transportation and distribution for phosphorus (P) in the downstream lake because external sources of P such as rainfall, runoff, soil leaching, industrial and municipal effluents ultimately reach lakes through the river–lake system. Distribution of P in the river is one of the key elements to limit downstream lake productivity.

In general, mineral phase solubility, redox reactions and mineralization of organic matter are the factors to control P release from sediments. It can be accessed through pH, temperature, dissolved oxygen, nitrates, sulfates and salinity (House & Denison 2002). Such release may have a significant impact on water quality and result in continuing eutrophication (Zhang et al. 2004). The P release from sediments is also affected by the changes in dry/wet cycles of sediments in river–floodplain systems (Schonbrunner et al. 2012). Since P has several fractions to distribute and/or migrate into the environment, its contribution in eutrophication promoting should thus be evaluated based on fraction share rather than content of total phosphorus (TP) alone (Tian & Zhou 2008; Withers & Jarvie 2008). Concerning the potential P-bioavailability, P can be classified into readily bioavailable, water-soluble and readily desorbable. According to the way of P fractionations, sedimentary P may be characterized as metal (Ca, Fe, Al) bound, occluded, labile, organic and reductant forms by using different extractants. Therefore, it is more efficient to evaluate the behavior of sediment-bound P on the base of different P fractions (Pardo et al. 1998).

The transport of P in rivers is usually dependent on its content in the water and sediments. The river sediments may act both as a sink or a source of P. In addition, P can release to the water from the sediments as an internal source for the river (House 2003; Evans et al. 2004). Both the move of sediment in rivers and the release of P from sediments are main controlling factors for the trophic status of the downstream lake (Kim et al. 2003). In particular, fine-grained suspended particulate (<63 μm) is an important transport vector of P in rivers (Stone et al. 1995).

Stream flow is key to controlling sediment transport in the river–lake system, which makes P retention in rivers highly variable in time and space. Sediments can deposit on the river bed in the low-flow period, and be removable during the flood event. Depending on grain size, water velocity and degree of turbulence, the concentration of particles in water will change both in the vertical and horizontal direction. In the past century, many dams and sluices have been built in watersheds to develop water resources and prevent flood disaster all over the world, which profoundly affect river hydrology, primarily through changes in the timing, magnitude and frequency of high and low flows, ultimately producing a hydrological regime that differs significantly from the pre-impoundment natural flow regime (Hu et al. 2008). Those hydraulic facilities regulate the stream flow and the transport of sediments, and thus control the behavior of P in river–lake systems. The relative importance of particulate versus dissolved delivery has decreased in the Mississippi River as a result of damming and fertilizer use in the watershed (Mayer et al. 1998). Hydraulic facilities such as weirs, impoundments, ponds and reservoirs connected to the river promote siltation and long-term particulate P retention because of much higher water residence times (Demars et al. 2005). However, the knowledge of in-stream sluices for P transport is still limited, although large pools of P retained in channels and behind sluices are subject to biogeochemical cycling processes.

In this study, the Huai River and Hongze Lake, a typical river–lake system in eastern China was chosen to study the spatial distributions of P in sediments of the Huai River catchment area where over 11,000 dams and sluices had been constructed along the general and tributary rivers for flood control since the 1960s (IGSNRR, WRBHRC 2006). The purpose of this paper was to examine the characteristics of P fractions in the sediments and the factors influencing their distribution along the river–lake system, giving particular attention to grain size and in-stream sluice.

MATERIALS AND METHODS

The characteristics of the Huai River catchment area

The Huai River catchment area (30°55′–36°36′N, 111°55′–121°25′E) is located between the Yangtze River catchment area and the Yellow River catchment area of China. It has 35 cities and 189 counties with a population of 168 million. The population density is about five times higher than the national mean of China (2005–2007). The Huai River is 1,000 km long and one of the most important rivers in China with a total drainage area of 270,000 km2. It flows through four provinces: Henan, Shandong, Jiangsu and Anhui. As the fourth biggest fresh lake in China, the Hongze Lake is mainly fed by the Huai River. The river and lake together form a typical river–lake ecosystem in China (Figure 1). In recent decades, the Hongze Lake has been seriously eutrophicated with microalgae blooms and red tides (Huang et al. 2010). With industrial and agricultural development and population growth upstream of the Huai River, large amounts of untreated wastewater and domestic sewage were released into Hongze Lake (Han et al. 2002; Huang et al. 2010).

Figure 1

Map of the Huai River Basin and sampling sites along the Huai River and the Hongze Lake.

Figure 1

Map of the Huai River Basin and sampling sites along the Huai River and the Hongze Lake.

The west side of the Huai River catchment area is generally situated at a higher elevation than the east side. The slope of the mainstream is 0.2 m/km in the upper reach between S1 and S3, then decreases to 0.1 m/km in the reach before the Bengbu Sluice (S6), and becomes a negative value of −0.01 m/km in the lower reach (S6–S9). However, the elevation of the bottom of the Hongze Lake (S10–S13) is about 10–11 m from sea level, about 3–4 m higher than that at S8 (Wang et al. 2000; Figure 1). It is a complex situation for water and sediment transport in the river–lake system. The catchment area is divided into mountainous areas, hills, plains and swales, consisting of 13%, 19%, 52% and 16% of the catchment area, respectively. Gneiss and migmatite are primarily distributed in the northwestern mountains, and schist in the western and southwestern mountainous areas; carbonates and petroliferous formations are widespread in the northern, central and eastern sections; unconsolidated sediments of tertiary and quaternary are distributed over the large plains.

The annual mean precipitation is 920 mm, more than 60% of which falls during the flood season from April to October. The annual mean temperature is 14°C. Because of climate and topography, the Huai River catchment area has been known for its frequent flood disasters. The major tributaries within the middle stream region lead to complexities in flood water routing. In addition, the East Line of the South to North Water Transfer Project becomes one of the most significant channels to supply water to the North China Plain through this region. To prevent flood disasters and develop water resources, about ten sluices or dams have been built, on average, at each tributary of the Huai River, with the total storage capacity of dams and sluices of 303 billion m3 in the catchment area. For example, the Bengbu Sluice (S6), a large-scale sluice, is the key facility in the middle reach of the Huai River to control water flow in the mainstream (Figure 1). It was built mainly for irrigation with accompanying functions for navigation and electricity generation, in 1962. As the largest in-stream sluice in the Huai River catchment area, the Bengbu Sluice consists of 30 check gates, most of which have a width of 10 m and a height of 7.5 m. It completely controls the flow and sediment in the reach, reducing runoff in dry season by 6%, but increasing flood flow in wet season by up to 47% because of the sluice operation. However, many contradictory arguments have been raised for many years regarding the benefits of building dams and sluices, and their detrimental impacts on the environment (Zhang et al. 2009; Wang & Xia 2010). In this paper, the study area is limited to the mainstream of the Huai River and the Hongze Lake, herein still referred to as the Huai River catchment area.

Sampling

The fieldwork was conducted in July 2008 to collect the surficial (10 cm) sediments throughout the river–lake system: nine samples (S1–S9) from the mainstream of the Huai River and four samples (S10–S13) from the Hongze Lake into which the Huai River flows (Figure 1). A large-scale sluice was constructed to control water flow at S6, and S9 was an inlet of the river before entering into the Hongze Lake. Moreover, the area from S10 to S11 was near the river inlet, and the area from S12 to S13 was away from the river inlet in the Hongze Lake. Sediment samples were collected using a Van Veen grab sampling device (Hydro-Bios Apparatebau GmbH, Kiel, Germany). The samples were put in air-sealed plastic bags and kept cool in the field (4°C). After transportation to the laboratory, they were kept frozen until analysis.

Experimental analysis

In order to analyze the P in the sediment, the samples were dried at room temperature. Sedimentary fractions of P are P-forms binding to metals and organic matter, such as exchangeable P (Ex-P), aluminum-bound P (Al-P), iron-bound P (Fe-P), occluded P (Oc-P), calcium-bound P (Ca-P), detrital-P (De-P) and organic P (Or-P). Ex-P is found mostly in the pore-water of sediments and released easily from sediments. Temperature, pH, hydrodynamic conditions and biological disturbance can cause Ex-P to diffuse up to overlying water. Al-P and Fe-P are exchangeable and represent the redox-sensitive P forms (Kaiserli et al. 2002) that can release into overlying water as soluble P when changing oxidation-reduction conditions. Oc-P is difficult to release from sediments because of the P covered by mineral particles of iron and aluminum. The P associated with calcium can be divided into two parts, De-P and Ca-P (Ruttenberg 1992). De-P is mainly derived from weathering products of the apatite mineral crystal dust. Ca-P represents self-generating, self-biogenic apatite P, and is deposited with self-generating calcium carbonate such as biological debris, fish bones and algae, etc. Both fractions are regarded as a permanent sink of P, just like Oc-P. Or-P mainly comes from the debris of aquatic microorganisms. Or-P in the sediment gradually mineralizes to form inorganic P, and is released into the water again.

The geographical distribution and forms of P in the sediments of the mainstream of Huai River and the Hongze Lake were examined by sequential extraction techniques. The sequential extraction scheme (Ruttenberg 1992; Kovar & Pierzynski 2009) with slight modifications (Li et al. 1998; Zhu et al. 2006) was used for P fractionation of the sediments. It was composed of seven steps: MgCl2-extractable for Ex-P; NH4F-extractable for Al-P; NaOH, Na2CO3-extractable for Fe-P; C6H5O7Na3, NaHCO3, Na2S2O4-extractable for Oc-P; NaAc-HAc, MgCl2-extractable for Ca-P; and HCl-extractable for De-P. The extraction of Or-P was performed with HCl after incubation at 550°C for 2 h. The extracts were centrifuged at 5,000 rpm for 20 min and the supernatants were filtered through 0.45 μm GF/C filter membrane.

The P contents in the extracts were measured by the molybdenum blue/ascorbic acid method (GB7852-87, State Standard of China) using a spectrophotometer (Beijing Tuopu Instrument Co., Ltd, TU1810, Beijing, China) adjusted to a wave-length of 700 nm. The P in sediments was expressed as mg/kg dry sediments. The total P of sediment (TPS) was calculated by summing up all of the seven fractions of P. In addition, the seven fractions of P could be regrouped to explain the regional characteristics of P patterns. In terms of biological uptake, P in sediments was divided into biologically available P (BAP) (=Fe-P + Al-P + Ex-P + Or-P) and non-biologically available P (NBAP) (=Ca-P + De-P + Oc-P). Furthermore, the sum of Fe-P, Al-P and Ex-P was defined as the inorganic biologically available P (IBAP) in this paper.

Sediments were analyzed using a laser diffraction particle size analyzer, and the grain sizes studied were classed according to the United States Department of Agriculture (USDA; Soil Survey Division Staff, 1993): clay (<0.002 mm), silt (0.002–0.05 mm), very fine sand (0.05–0.125 mm), fine sand (0.125–0.25 mm), medium sand (0.25–0.50 mm) and coarse sand (0.50–2.00 mm).

The pH of sediments were measured in a deionized water suspension (1:2.5 v/v) using a pH meter in laboratory conditions.

RESULTS

Spatial distribution of grain size characteristics and pH

Sediments in the Huai River catchment area occur as mixtures with varying proportions of particles of different sizes. Each component contributes its characteristics to the mixture. Figure 2 shows the variations of sediment components throughout the river–lake system. Clay made the lowest contribution with a mean of 1.6% (0.0–2.4% range) in the sediments. Silt, with a mean of 53.8% (2.5–77.6% range) was the dominant part of sediments in the river–lake system, excepting S1. The total amount of sands had a mean of 44.6% (20.1–97.5% range). Very fine sand in sediments with a mean of 19.8% (2.1–28.6% range) had the lowest value at S1 and became the dominant part of sand from S4 to S13. Fine sand, with a mean of 15.9% (1.3–38.7% range), was the major component of sand from S1 to S3 and the second dominant sand from S4 to S13. Medium sand was the dominant sand with the highest value of 48.1% at S1. It decreased continually through the river–lake system and was lower than 10% from S4 to S13. Coarse sand was less than 1.0% from S3 to S13, and the smallest part of sand in the river system.

Figure 2

Grain size characteristics along the Huai River and the Hongze Lake.

Figure 2

Grain size characteristics along the Huai River and the Hongze Lake.

The mean pH value of sediments was 9.1 (8.0–10.4 range) in the mainstream of the Huai River and 9.4 (8.6–10.0 range) in the Hongze Lake (Table 1). On average, the pH of sediments in the mainstream was lower than that in the lake.

Table 1

Standard deviation and variations of pH, TPS and P fractions in the sediments of the mainstream of the Huai River and the Hongze Lake

  Mainstream
 
Hongze Lake
 
Items Min. Max. Mean S.D. Min. Max. Mean S.D. 
pH 8.0 10.4 9.1 0.6 8.6 10.0 9.4 0.6 
TPS (mg/kg) 518.1 1424.2 881.3 248.5 661.0 824.3 823.1 62.5 
Oc-P (mg/kg) n.d. 2.3 1.3 1.0 n.d. 1.9 0.9 0.9 
Ca-P (mg/kg) 20.0 214.5 75.5 62.3 9.3 69.2 38.8 25.4 
De-P (mg/kg) 187.9 1181.8 471.1 284.1 178.8 575.5 374.7 178.9 
Al-P (mg/kg) n.d. 22.7 13.4 9.6 n.d. 25.3 12.3 12.3 
Ex-P (mg/kg) 8.9 91.2 51.2 23.1 44.0 58.2 50.3 5.8 
Fe-P (mg/kg) 22.8 310.4 128.3 94.1 64.3 140.9 109.1 29.8 
Or-P (mg/kg) 44.9 344.7 140.6 79.8 89.4 369.6 237.0 131.1 
  Mainstream
 
Hongze Lake
 
Items Min. Max. Mean S.D. Min. Max. Mean S.D. 
pH 8.0 10.4 9.1 0.6 8.6 10.0 9.4 0.6 
TPS (mg/kg) 518.1 1424.2 881.3 248.5 661.0 824.3 823.1 62.5 
Oc-P (mg/kg) n.d. 2.3 1.3 1.0 n.d. 1.9 0.9 0.9 
Ca-P (mg/kg) 20.0 214.5 75.5 62.3 9.3 69.2 38.8 25.4 
De-P (mg/kg) 187.9 1181.8 471.1 284.1 178.8 575.5 374.7 178.9 
Al-P (mg/kg) n.d. 22.7 13.4 9.6 n.d. 25.3 12.3 12.3 
Ex-P (mg/kg) 8.9 91.2 51.2 23.1 44.0 58.2 50.3 5.8 
Fe-P (mg/kg) 22.8 310.4 128.3 94.1 64.3 140.9 109.1 29.8 
Or-P (mg/kg) 44.9 344.7 140.6 79.8 89.4 369.6 237.0 131.1 

S.D.: standard deviation; n.d.: below detection limit.

Variations of TPS and P fraction concentration

Table 1 shows the variations of TPS and P fractions in the mainstream of the Huai River and the Hongze Lake. Based on the data from sediment samples, the mean TPS concentration of the river–lake system was 863.4 mg/kg, with a range of 518.1–1424.2 mg/kg. The standard deviation (S.D.) was 212.7. Comparing between the mainstream and the lake, the mean of TPS in the mainstream was higher than that in the lake. In terms of P fractions, the mean contents of Ca-P, De-P and Fe-P showed similar variation trends as the mean of TPS between the mainstream and the lake, respectively, but the mean of Or-P displayed a contrary tendency. The mean contents of Oc-P, Al-P and Ex-P were similar through the river–lake system. As a whole, the rank order of P fractions was: De-P > Or-P > Fe-P > Ca-P > Ex-P > Al-P > Oc-P in the mainstream of the Huai River, and De-P > Or-P > Fe-P > Ex-P > Ca-P > Al-P > Oc-P in the Hongze Lake.

Tables 2 and 3 show the Pearson correlation matrix for pH, TPS and P fractions of the mainstream of the Huai River and the Hongze Lake, respectively. Comparatively, Oc-P was the only P form which had significant positive correlations with Ca-P and Al-P in both the river and lake. However, significant correlations among other P forms were different in the river and lake. For example, Ca-P in sediments had a significant correlation with TPS in the mainstream of the Huai River, but with De-P in the Hongze Lake. Al-P in sediments had significant correlations with pH and TPS, Ex-P in the river, but Ca-P and De-P in the lake. Fe-P in sediments had only a negative significant correlation with De-P in the river, and no significant correlation with any other forms of P, or TPS of sediment. Or-P in sediments had no significant correlations with any P fractions in the river, but significant negative correlations with Al-P, Oc-P, Ca-P and De-P in the lake. In addition, pH had positive correlations with Oc-P, Al-P and Ex-P in the river, but no significant correlation with any other forms of P, or TPS of sediment in the lake.

Table 2

Pearson correlation matrix for pH, TPS and P fractions of the mainstream of the Huai River

 TPS Oc-P De-P Ca-P Al-P Ex-P Fe-P Or-P 
pH 0.66 0.84a −0.43 0.57 0.85a 0.67b 0.25 −0.24 
TPS  0.83a −0.30 0.94a 0.71b 0.23 0.15 −0.34 
Oc-P   −0.41 0.73b 0.91a 0.53 0.44 −0.51 
De-P    −0.14 −0.30 −0.53 −0.69b −0.19 
Ca-P     0.64 0.06 −0.11 −0.50 
Al-P      0.71b 0.33 −0.54 
Ex-P       0.59 −0.16 
Fe-P        0.00 
 TPS Oc-P De-P Ca-P Al-P Ex-P Fe-P Or-P 
pH 0.66 0.84a −0.43 0.57 0.85a 0.67b 0.25 −0.24 
TPS  0.83a −0.30 0.94a 0.71b 0.23 0.15 −0.34 
Oc-P   −0.41 0.73b 0.91a 0.53 0.44 −0.51 
De-P    −0.14 −0.30 −0.53 −0.69b −0.19 
Ca-P     0.64 0.06 −0.11 −0.50 
Al-P      0.71b 0.33 −0.54 
Ex-P       0.59 −0.16 
Fe-P        0.00 

aCorrelation is significant at the 0.01 level (two-tailed).

bCorrelation is significant at the 0.05 level (two-tailed).

Table 3

Pearson correlation matrix for pH, TPS and P fractions of the Hongze Lake

 TPS (mg/kg) Oc-P De-P Ca-P Al-P Ex-P Fe-P Or-P 
pH −0.26 0.99 0.87 0.97 0.93 0.81 −0.62 −0.90 
TPS  −0.32 0.03 −0.22 −0.13 −0.56 −0.21 0.07 
Oc-P   0.94 0.99a 0.98b 0.39 −0.81 −0.97b 
De-P    0.97b 0.98b 0.21 −0.93 −0.99a 
Ca-P     0.99b 0.37 −0.84 −0.98b 
Al-P      0.23 −0.91 −1.00a 
Ex-P       0.17 −0.20 
Fe-P        0.93 
 TPS (mg/kg) Oc-P De-P Ca-P Al-P Ex-P Fe-P Or-P 
pH −0.26 0.99 0.87 0.97 0.93 0.81 −0.62 −0.90 
TPS  −0.32 0.03 −0.22 −0.13 −0.56 −0.21 0.07 
Oc-P   0.94 0.99a 0.98b 0.39 −0.81 −0.97b 
De-P    0.97b 0.98b 0.21 −0.93 −0.99a 
Ca-P     0.99b 0.37 −0.84 −0.98b 
Al-P      0.23 −0.91 −1.00a 
Ex-P       0.17 −0.20 
Fe-P        0.93 

aCorrelation is significant at the 0.01 level (two-tailed).

bCorrelation is significant at the 0.05 level (two-tailed).

Classification of P patterns based on its fractions

To elaborate on the characteristics of P fractions in the study area, a comprehensive diagram has been developed (Figure 3). It consists of three individual diagrams, two triangles and one diamond. The triangle on the right stands for the components of BAP and that on the left for components of NBAP. The diamond in the middle comprehensively displays the characteristics of seven P fractions. The relative abundance of BAP with the fractions of Ex-P (Fe-P + Al-P) and Or-P assumed to equal 100% is first plotted on the BAP triangle. Similarly, the NBAP triangle displays the relative abundance of Ca-P, De-P and Oc-P. The straight lines projecting from the two triangles into the quadrilateral field define the position of the point on the third field. As a result, the sediments were classified according to fractions of P with two templates for the diagram, one for the BAP and the other for the NBAP. The limited number of possibilities for classifying the P data effectively eliminated local variability and preserves broad trends.

Figure 3

Diagrams for phosphorus characteristics of sediments in the Huai River and the Hongze Lake.

Figure 3

Diagrams for phosphorus characteristics of sediments in the Huai River and the Hongze Lake.

BAP in TPS in the Hongze Lake (51.1%) was higher than that in the mainstream of the Huai River (40.0%). However, IBAP is similar in both the river and lake, ranging from 21.0 to 22.1% of TPS. Accordingly, BAP in TPS was about two times that of IBAP in TPS in the river–lake system. In the mainstream, De-P was the dominant P fraction of NBAP (84.1% of NBAP). Or-P (44.4% of BAP) was the dominant P fraction of BAP followed by (Fe-P + Al-P; 39.6% of BAP) in the sediments. The dominating P fractions of NBAP and BAP in the lake were De-P and Or-P, respectively. Accordingly, the P pattern in sediments of the mainstream changed from (De-P + Ca-P) + Or-P type in the upstream of Bengbu Sluice to De-P + Fe-P type in the downstream. De-P + (Fe-P + Al-P) type (S10 and S11) and De-P + Or-P type (S12 and S13) were found in the lake. As a result, De-P was the dominant fraction of NBAP in the Huai River catchment area. However, the relative abundance of Or-P in BAP ranged from 19.5 to 80.5% and decreased from the upstream to downstream in the mainstream. It had a range of 39.8–67.9% in the lake with a decreasing trend lakeward. Fe-P was another important component in BAP. Its mean percentage in BAP was 35.1% in the mainstream and 27.7% in the lake. Therefore, BAP components are those fractions relating to absorption capacity of the sediment, the redox condition as well as biological activity in the river–lake system.

In addition, the direction of river flow is shown in the diamond diagram where the size of the data point is related to TPS to provide an indication of the absolute quantity in the sample. It was found that TPS increased continuously along the mainstream (Figure 3).

Variations of P fractions throughout the river–lake system

Because the regulation practices of the sluice could affect transport and distribution of P in the reach, it was reasonable to divide the reach into three parts, including the upper reach from S1 to S5 in front of Bengbu Sluice (S6), the lower reach from S6 to S9, and the lake (S10–S13).

Figure 4 shows the variations of P fractions in the river–lake system. Both BAP and NBAP had trends of increasing along the river in the upper reach. After it flowed downward to the Bengbu Sluice, NBAP increased continuously in the reach between S6 and S8, while BAP decreased at the same reach. Thus, S8 became the turning point with the highest value for NBAP and the lowest value for BAP. BAP in the Hongze Lake returned to a level similar to that in the reach between S4 and S5. NBAP in the lake finally returned to a low level of about 234.0 mg/kg at S13. As a whole, BAP components changed largely between the upstream and downstream of Bengbu Sluice.

Figure 4

Variations of P fractions in the river–lake system of the Huai River catchment area.

Figure 4

Variations of P fractions in the river–lake system of the Huai River catchment area.

De-P had a similar variation trend to NBAP in the river–lake system. However, the content of Ca-P with the mean percentage of 13.4% in NBAP displayed a decreasing trend in both the upper and lower reaches. Ca-P in the lake sediments showed levels at S10, S11 similar to that in the lower reach, except S6, and a decreasing trend at S12 and S13. In addition, Oc-P as one of the steady trophic states of P had the lowest content, accounting for less than 0.2% of NBAP, and its contribution to the eutrophication was limited in the studied river–lake system.

The P fractions of BAP appeared to have different variation trends throughout the river–lake system. The content of Or-P as the dominant P fractions of BAP showed an increasing trend in the upper reach, and increased continuously until S6 (140.0 mg/kg). However, Or-P reduced significantly at S7 (44.9 mg/kg), and after that, it increased again toward S9 (157.7 mg/kg). Or-P in the lake sediments at S12 and S13 (369.6–365.4 mg/kg range) was about three to four times higher than that at S10 and S11. Fe-P and Ex-P, with the mean percentage of 32.8% and 15.5% in BAP, respectively, had a similar increasing trend to Or-P in both of the upper and lower reaches of the mainstream. Nevertheless, S6 became the turning point of content variation. In the lake, Fe-P at S12 and S13 were higher than that at S10 and S11. However, the variation of Ex-P was relatively stable. Al-P with the mean percentage of 4.6% in BAP was not detectable from the upstream of the Huai River until reaching S4 (about 343 km from the S1), and ranged from 16.8 to 22.7 mg/kg in the reach between S4 and S9. After it flowed into the lake, Al-P was as high as 24.0 to 25.3 mg/kg in the area near the inlet of the river (S10 and S11) and below detection limits at S12 and S13.

DISCUSSION

Relationship between P fractions and grain size characteristics

Significant correlations with certain grain size fractions were found for Fe-P, Ex-P and Ca-P. Weathering in the watershed has implications for mechanisms of P incorporation into the sediments and their availability for inorganic release processes and uptake by biota (House & Denison 2002). Fe-P had a significant positive correlation with very fine sand. Although Ex-P and Ca-P had good relationships with all size fractions, their characteristics were different from each other. Ex-P was positively correlated with fine grain (<0.125 mm) and negatively with large grain (>0.125 mm). It was therefore subject to more complex controls than those exerted by sediment mineralogical composition alone, particularly in flowing waters where microbial and algal biofilms, hydrodynamic factors, bioturbation, redox conditions and pH played a critical role in Ex-P at the sediment water interface (House 2003). However, Ca-P was positively correlated with large grain and negatively with fine grain. It seemed that large size particles often exhibited high concentrations of Ca-P. In addition, the distribution pattern of De-P over the size classes had no marked feature, suggesting that surface effects were of minor importance and that calcium phosphate was mainly present within the matrix of the sediment minerals. TPS was positively correlated with clay and silt and negatively with fine sand. BAP was only negatively correlated with fine sand and had less significant correlation with other grain sizes. Without considering the Or-P, IBAP positively correlated with very fine sand and negatively with medium sand. In general, Oc-P, Or-P, Al-P and NBAP had no significant corrections with grain size characteristics. Although the relations between P-forms and grain size characteristics were significant or weak, a grain size of 0.125 mm seemed to be a threshold value (Table 4). As a whole, most P-forms had positive correlation coefficients below the threshold size and negative ones above the value while Ca-P had an inverted behavior.

Table 4

Pearson correlation matrix for the phosphorus fractions, total organic carbon and grain size characteristics

 Clay Silt Very fine sand Fine sand Medium sand Coarse sand 
Oc-P 0.39 0.42 0.41 −0.38 −0.45 −0.44 
Ca-P −0.66b −0.72a −0.62b 0.63b 0.76a 0.60b 
De-P 0.33 0.32 −0.01 −0.24 −0.24 −0.22 
Al-P 0.33 0.35 0.40 −0.26 −0.43 −0.48 
Ex-P 0.61b 0.64b 0.72a −0.59 −0.71a −0.70a 
Fe-P 0.30 0.38 0.62b −0.46 −0.43 −0.34 
Or-P 0.28 0.28 0.01 −0.38 −0.11 0.09 
TPS 0.56b 0.57b 0.15 −0.56b −0.39 −0.28 
BAP 0.49 0.54 0.49 −0.65b −0.46 −0.26 
NBAP 0.19 0.016 −0.14 −0.10 −0.07 −0.09 
IBAP 0.41 0.49 0.71a −0.54 −0.56b −0.48 
TOC 0.40 0.41 −0.29 −0.52 −0.03 0.18 
 Clay Silt Very fine sand Fine sand Medium sand Coarse sand 
Oc-P 0.39 0.42 0.41 −0.38 −0.45 −0.44 
Ca-P −0.66b −0.72a −0.62b 0.63b 0.76a 0.60b 
De-P 0.33 0.32 −0.01 −0.24 −0.24 −0.22 
Al-P 0.33 0.35 0.40 −0.26 −0.43 −0.48 
Ex-P 0.61b 0.64b 0.72a −0.59 −0.71a −0.70a 
Fe-P 0.30 0.38 0.62b −0.46 −0.43 −0.34 
Or-P 0.28 0.28 0.01 −0.38 −0.11 0.09 
TPS 0.56b 0.57b 0.15 −0.56b −0.39 −0.28 
BAP 0.49 0.54 0.49 −0.65b −0.46 −0.26 
NBAP 0.19 0.016 −0.14 −0.10 −0.07 −0.09 
IBAP 0.41 0.49 0.71a −0.54 −0.56b −0.48 
TOC 0.40 0.41 −0.29 −0.52 −0.03 0.18 

aCorrelation is significant at the 0.01 level (two-tailed).

bCorrelation is significant at the 0.05 level (two-tailed).

For the organic fraction, it is mainly associated with the fine particles of sediments. Organic matter in the sediments had the mean of 7.5% (3.9–11.17% range) in the area of the river inlet (S10–S11), 4.8% (2.1–7.4% range) in the west side of the lake, and 1.2% (0.9–1.5% range) in the area away from the inlet (S12–S13) (He et al. 2005). For sediments in the Huai River catchment area, total organic carbon (TOC) had a weak positive relation with silt and clay. We also examined the relationship between TOC and Or-P and found that the correlation between them was not significant. The land-use practices and geology affect P limitation of stream sediment organisms (Klotz 1985).

Variation of P distribution in the river–lake system associated with the Bengbu Sluice

The impact of hydrological engineering on river flow, sediments and associated nutrients is an important topic both for the catchment area and for wider environmental protection. Flow regulation has been widely applied to many rivers in China. For example, flow in the mainstream as well as most tributaries of the Huai River has been regulated to reduce the risks of floods and smooth the yearly runoff in the catchment area (Wang & Xia 2010). As a result, the sluices in the mainstream of the Huai River control floods by cutting off river flow and slowing down the flow velocity in the upper and downstream of dams and sluices (Zhang et al. 2009). The Bengbu Sluice affected water quality behind it by increasing the concentration of ammoniacal nitrogen (NH4-N) and chemical oxygen demand (CODMn) (Zhang et al. 2009).

Phosphorus may be absorbed to particles with a fine size such as clay, silt and very fine sand. Grain size is a decisive factor in the processes of transport and sedimentation. In terms of sediment transport in the Huai River and Hongze Lake system, 93.0% of the influx of sediments for the lake came from the Huai River, based on the data from 1975 to 2004 as to the effect of dams and sluices in the catchment area (Chen et al. 2009). The sediment load has seasonal changes in the channel. For example, the operation of Bengbu Sluice lets the sediments deposit at the reach during the non-flood period and releases the sediment to the downstream. However, the sediment load at S6 was similar to that at S8 all year round, showing that the shape of river bed remained constant in the reach (Yu et al. 2009). Based on grain size data in this paper, a variation was showed between the upper reach and the lower reach of the river–lake system. The sum of very fine sand and fine sand in sediments in the reach downstream of the Bengbu Sluice was less than that in the reach upstream as well as that in the Hongze Lake (Figure 2). Sand grains of 0.2 mm size are most easily moved because of the roughness velocity, settling velocity and threshold velocity. In addition, since both finer and coarser sediments are more difficult to move, bottom sediments in the act of transport become better sorted as the diameter approaches 0.2 mm (Sanders 1958).

The Bengbu Sluice not only affected distribution of the sediment grain size, but also distribution of P fractions. Figure 5 shows the mean contents of P fractions in the river–lake system. De-P was the dominant P form distributed widely in the sediments as the base of P, and the other P forms are the result of fluvial function in the river–lake system. Except for Fe-P and Or-P, the means of P fractions in the reach downstream of the Bengbu Sluice were higher than both the reach upstream of the sluice and the lake. Fe-P and Or-P were easily transported by suspended load and the difference of sediment load was small in the reach between S6 and S9. Fe-P and Or-P in the reach downstream of the Bengbu Sluice were lower than both the reach upstream and the lake, because of the difficulty for them of depositing in the reach. Despite large changes of P fractions in the reach downstream of the Bengbu Sluice, the means of Oc-P, De-P and Ex-P were similar in both the reach upstream of the Bengbu Sluice and the lake. The mean of Ca-P in the lake sediments decreased to 38.8 mg/kg, half of that in the mainstream. The mean of Fe-P in the lake was only two-thirds of that in the reach upstream of the Bengbu Sluice. The mean of Or-P was higher than that in the mainstream, indicating that Or-P was easier to deposit in the lake environment than other P fractions. Therefore, it seemed that the Bengbu Sluice was a point to regulate transport of P in sediment upstream and downstream of the mainstream, as well as the distribution of P in the river–lake system.

Figure 5

Mean contents of phosphorus fractions in different parts of the Huai River and the Hongze Lake.

Figure 5

Mean contents of phosphorus fractions in different parts of the Huai River and the Hongze Lake.

CONCLUSIONS

This study evaluated characteristics of P fractions in sediments and the effect of grain size and in-stream sluice in a river–lake system.

  1. The mean of TPS and P fractions in the mainstream of the Huai River was higher than that in the Hongze Lake, except for Fe-P and Or-P.

  2. In terms of the dominant fractions of BAP and NBAP, the P pattern in sediments changed from (De-P + Ca-P) + Or-P type at the reach upstream of the Bengbu Sluice to De-P + Fe-P type at the downstream. It was also identified as De-P + (Fe-P + Al-P) type and De-P + Or-P type in the Hongze Lake.

  3. Fe-P, Ex-P and Ca-P had significant relationships with grain size characteristics but not with other P fractions. The grain size of 0.125 mm could be considered as the threshold value to assess the P fractions in the study area.

  4. The great variations of P fractions upstream and downstream of Bengbu Sluice indicated the in-stream sluice was also important to assess P from the point of view of fluvial geomorphology.

Therefore, further research is needed to explore management options to reduce P and enhance the natural retention capacity of streams using comprehensive engineering solutions, such as integrated sluice operation, to increase the range of habitats and flow velocities in streams, preventing nutrient loadings from reaching the watershed, as well as improving ecological diversity.

ACKNOWLEDGEMENTS

This work was funded by the Main Direction Program of Knowledge Innovation of the Chinese Academy of Sciences (no. KZCX2-YW-Q06-1). We thank our laboratory colleagues for their collaboration in sampling and data acquisition. We would like to express our appreciation to Prof. Chen Jianyao from Sun Yatsen University, China, for analyzing grain size characteristics.

REFERENCES

REFERENCES
Chen
L.
Zhang
W.
Yu
H.
Lu
S.
2009
Study on floating sand accumulation, physicochemical characteristics and heavy metal pollution in the sediment of Lake Hongze
.
Chinese Agricultural Science Bulletin
25
,
219
226
.
Demars
B.
Harper
D. M.
Pitt
J.-A.
Slaughter
R.
2005
Impact of P control measures on in-river P retention associated with point source pollution
.
Hydrology and Earth System Sciences Discussions
2
,
37
72
.
Evans
D. J.
Johnes
P. J.
Lawrence
D. S.
2004
Physico-chemical controls on P cycling in two lowland streams. Part 2 – the sediment phase
.
Science of the Total Environment
329
,
165
182
.
GB7852-87
1987
State Standard of China for Determination of Total Phosphorus in Forest Soil
.
Standardization Administration of the People's Republic of China
,
Beijing, China
.
Han
A.
Yang
G.
Liu
B.
Zhang
S.
2002
Investigation and evaluation about the state of eutrophication and ecology of the lake of Hongze
.
The Administration and Technique of Environmental Monitoring
14
,
18
20
(in Chinese)
.
He
H.
Ding
H.
Zhang
Z.
Shi
X.
Li
S.
Mao
L.
2005
Grain size characteristics and their environmental significance of Hongze Lake sediments
.
Scientia Geographica Sinica
25
,
590
596
.
House
W. A.
2003
Geochemical cycling of P in rivers
.
Applied Geochemistry
18
,
739
748
.
Huang
L.
Sun
K.
Ban
J.
Bi
J.
2010
Public perception of blue-algae bloom risk in Hongze Lake of China
.
Environmental Management
45
,
1065
1075
.
IGSNRR, WRBHTC
2006
Impacts of Dams and Sluices on Ecology and Environment in HRB
.
Institute of Geographic Sciences and Natural Resources Research
,
Beijing, China
.
Klotz
R.
1985
Factors controlling P limitation in stream sediments
.
Limnology and Oceanography
30
,
543
553
.
Kovar
J. L.
Pierzynski
G. M.
2009
Methods of P Analysis for Soils, Sediments, Residuals, and Waters. Southern Cooperative Series Bulletin No. 408
.
North Carolina State University Raleigh
, p.
122
.
Li
Y.
Wu
D.
Xue
Y.
1998
A development sequential extraction method for different forms of P in the sediments and its environmental geochemical significance
.
Marine Environmental Science
17
,
15
20
.
Mayer
L. M.
Keil
R. G.
Macko
S. A.
Joye
S. B.
Ruttenberg
K. C.
Aller
R. C.
1998
Importance of suspended participates in riverine delivery of bioavailable nitrogen to coastal zones
.
Global Biogeochemical Cycles
12
,
573
579
.
Sanders
H.
1958
Benthic studies in Buzzards Bay. I. Animal-sediment relationships
.
Limnology
3
,
245
258
.
Schonbrunner
I. M.
Preiner
S.
Hein
T.
2012
Impact of drying and re-flooding of sediment on P dynamics of river-floodplain systems
.
Science of the Total Environment
432
,
329
337
.
Soil Survey Division Staff
1993
Soil Survey Manual
.
United States Department of Agriculture
.
Wang
Q.
Li
J.
Dao
J.
Chen
J.
2000
Formation, evolution and harnessing of antislope river bed along the middle Huaihe River
.
Journal of Sediment Research
1
,
50
55
.
Withers
P. J.
Jarvie
H. P.
2008
Delivery and cycling of P in rivers: a review
.
Science of the Total Environment
400
,
379
395
.
Yu
B.
Yu
Y.
Ni
J.
2009
Analysis on erosion and deposition in Wujiadu-Xiaoliuxiang reach of the mainstream Huaihe River
.
Journal of Sediment Research
8
,
12
16
.
Zhang
L.
Fan
C.
Chi
Q.
Wang
J.
Qin
B.
2004
P species distribution of sediments in Lake Taihu and its inflow rivers
.
Geochimica
33
,
423
432
.