ABSTRACT Seasonal variation in phosphorus release potential from bottom sediment and environmental factors affecting phosphorus mobilization in shallow eutrophic lakes, Lakes Nishiura and Kitaura in Japan, were investigated. At uplake sites, the fluxes were low and showed no seasonal variation. The Eh in the sediments at these sites was positive and in an oxidizing state, whereas sediments at other sites were in a reducing state. The oxidation/reduction states were not affected by the presence of DO in the boundary layer water. DO in the boundary layer only oxidized the surface of sediment layer and affected DIP release from the sediment. The oxidation of sediment was due to nitrate input from inflowing rivers. At mid-lake sites, where the DIP release flux was high, the Eh in sediments was increased by nitrate input. Nitrate, which is an electron acceptor in the boundary layer, diffused into the sediment and oxidized the sediment layer in contrast to the definitive oxidation attributed to DO. The DIP release flux decreased with increasing nitrate supply rates. The sediment record shows that the oxidation state was maintained by nitrate supply, which is affected by the decomposition rate of nitrate and strongly affects DIP release from lake sediments under anaerobic conditions.
Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 163 - Effect of Nitrate on Phosphorus Mobilization from Bottom Sediment in Shallow Eutrophic Lakes Yuichi ISH I I* ,† , Tohru YABE**, Masako NAKAMURA***, Yoshimasa AMANO** ,†† , Nobuyuki KOMATSU* ,††† , Keiji WATANABE* * Water Environmental Laboratory, Ibaraki Kasumigaura Environmental Science Center, 1853 Okijuku, Tsuchiura, Ibaraki, 300-0023, Japan ** Environmental Biology Division, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki, 305-8506, Japan *** The United Graduate School of Agricultural Sciences, Tottori University, 4-101 Koyamacho-minami, Tottori, 680-8553, Japan † Present address: Environmental Biology Division, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki, 305-8506, Japan †† Present address: Graduate School of Engineering, Chiba University, 1-33 Yayoicho, Inage, Chiba, 263-8522, Japan ††† Present address: Fisheries Administration Division, Ibaraki Prefectural Government, 978-6 Kasahara, Mito, Ibaraki, 310-8555, Japan ABSTRACT Seasonal variation in phosphorus release potential from bottom sediment and environmental factors affecting phosphorus mobilization in shallow eutrophic lakes, Lakes Nishiura and Kitaura in Japan, were investigated. At uplake sites, the fluxes were low and showed no seasonal variation. The Eh in the sediments at these sites was positive and in an oxidizing state, whereas sediments at other sites were in a reducing state. The oxidation/reduction states were not affected by the presence of DO in the boundary layer water. DO in the boundary layer only oxidized the surface of sediment layer and affected DIP release from the sediment. The oxidation of sediment was due to nitrate input from inflowing rivers. At mid-lake sites, where the DIP release flux was high, the Eh in sediments was increased by nitrate input. Nitrate, which is an electron acceptor in the boundary layer, diffused into the sediment and oxidized the sediment layer in contrast to the definitive oxidation attributed to DO. The DIP release flux decreased with increasing nitrate supply rates. The sediment record shows that the oxidation state was maintained by nitrate supply, which is affected by the decomposition rate of nitrate and strongly affects DIP release from lake sediments under anaerobic conditions. Keywords: nitrate; phosphorus release; redox potential. INTRODUCTION Phosphorus is a key element affecting primary production in lakes (Vähälä et al., 2001; Kisand, 2005) and it is well known that lake sediments can act as a source or sink for phosphorus (Moore et al., 1998). This phosphorus is utilized by phytoplankton and plants and cycled within aquatic ecosystems. Metabolic byproducts of aquatic organisms and/or detritus material are decomposed by microbes, resulting in the release of the phosphorus accumulated in these organisms into the aquatic environment. Some of this matter and unutilized phosphorus sink and form sediments; most of the phosphorus in aquatic environments is accumulated in this way. Phosphorus release flux from sediments can contribute to the eutrophication process in lakes (Lennox, 1984). Phosphorus release is influenced by a variety of environmental factors, including water temperature, pH, form of phosphorus, dissolved oxygen (DO), nitrate, redox potential, and hydrological conditions (Jensen and Andersen, 1992; Kleeberg and Kozerski, 1997; Address correspondence to Yuichi ISHII, National Institute for Environmental Studies Email: ishii.yuichi@nies.go.jp Received 30 January 2009, Accepted 21 May 2009 Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 164 - Japan 36º E 140º N 0 5 10km0 5 10km N E W S L.Kitaura L.Nishiura N1 N4 N7 N8 K1 K3 K6 L.Sotonasakaura T o m o e R . H o k o t a R . Koise R. S a k u r a R . H i t a c h i t o n e R . W a n i R . Fig.1 – Location of study sites with sampling points. Gao et al., 2005; Jiang et al., 2006; Jin et al., 2006; Zhu et al., 2007). Phosphorus distribution in lake sediments is not uniform over an entire lake and environmental conditions are equally variable. Consequently, the release of phosphorus, and the factors affecting this release, is likely to vary within a lake. Lakes Nishiura and Kitaura are located in the eastern region of central Japan (Fig.1). There has been an increase in the inflow of domestic and industrial wastewater associated with the increase in urbanization and industrialization in these watersheds, resulting in increased eutrophication in the lakes. Nutrient levels are high in the central areas of Lake Kitaura, especially dissolved inorganic phosphorus (DIP), which has been observed to increase during the high water temperature period (Ishii and Komatsu, 2006). In fact, the highest DIP concentration of 334 μg l -1 was attributed to phosphorus release (Ishii et al., 2007). In addition, the annual increases observed in total phosphorus (TP) concentrations in these lakes (Ishii and Komatsu, 2006) mean that a detailed examination of the elevated phosphorus concentrations is required in order to prevent the occurrence of serious aquatic problems such as eutrophication. In this study, the release flux of DIP from lake sediments was investigated using sediment–water column samples collected in different seasons from various sites in the shallow eutrophic waters of Lakes Nishiura and Kitaura. The primary purposes of this study were to (i) examine the seasonal variation of phosphorus release potential from bottom sediments in these two lakes, and (ii) clarify the environmental factors effecting phosphorus mobilization in these lakes. MATERIALS AND METHODS Study site description Lakes Nishiura and Kitaura were formerly lagoons that were flooded by a rise in sea Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 165 - Tabl e 1 – Main characteristics of sampling points and physicochemical properties of sediment samples in Lakes Nishiura and Kitaura. Location Water depth (m) Water quality a Ave. (Min.-Max.) Ave. (Min.-Max.) pH 7.7 (7.0-8.7) 7.7 (7.2-8.0) 7.8 (7.2-8.2) 8.1 (7.4-8.7) 7.5 (7.0-8.0) 7.8 (7.4-8.2) 8.1 (7.7-8.6) EC (µS cm -1 ) 228 (180-266) 286 (271-303) 307 (286-355) 319 (300-352) 316 (270-359) 285 (267-294) 324 (289-373) DO (µg l-1) 6.7 (4.8-8.7) 5.1 (2.6-7.0) 5.5 (4.0-7.2) 6.7 (6.2-7.2) 7.2 (5.9-8.5) 5.1 (0.9-8.2) 7.2 (4.5-8.9) PO 4 -P (µg l -1 ) 19 (4-54) 10 (1-28) 11 (1-30) 6 (ND c -20) 15 (2-28) 46 (2-131) 31 (2-107) DOP (µg l-1) 10 (3-16) 7 (ND c -10) 7 (4-11) 9 (5-15) 8 (4-15) 6 (1-9) 8 (4-11) NO 3 -N (µg l -1 ) 1419 (281-2340) 48 (ND c -223) 9 (ND c -40) 9 (ND c -75) 4698 (2995-7008) 393 (30-980) 15 (3-46) NO 2 -N (µg l -1 ) 33 (19-46) 10 (1-40) 8 (1-30) 2 (ND c -8) 250 (122-420) 49 (3-100) 4 (ND c -21) NH 4 -N (µg l -1 ) 117 (16-341) 227 (17-813) 252 (14-525) 51 (7-170) 428 (65-1356) 192 (16-425) 20 (10-46) DON (µg l-1) 500 (371-655) 573 (394-737) 547 (400-660) 484 (387-640) 692 (391-1473) 516 (375-716) 567 (454-717) Sediment b Porosity (%) Water Content (%) Loss on Ignition (%) Mean Grain Size (µm) Clay (<3.9µm) (%) Silt (3.9-62.5µm) Very Fine Sand (62.5-125µm) Fine Sand (125-250µm) Medium Sand (250-500µm) a measured in this study b Ishii et al . (2008) c not detected --0.1---- 0.4 <0.1 43.3 ---- --15.0 0.4 - <0.1 0.6 36.4 58.2 10.9 63.9 45.5 43.1 59.3 63.2 41.8 30.7 35.8 54.5 56.9 40.1 22.2 21.2 6.1 6.1 3.0 2.8 5.4 6.0 2.0 38.1 20.5 18.4 18.8 10.6 42.9 44.2 45.5 65.4 72.6 74.0 57.9 68.9 86.1 47.6 48.7 45.5 46.8 45.5 uplake middlelake downlake 2.2 5.1 5.7 4.4 2.0 7.3 3.1 uplake middlelake middlelake downlake L. Nishiura L. Kitaura N1 N4 N7 N8 K1 K3 K6 level and subsequent closure of their respective bay mouths by sand dune when the sea level dropped to near existing levels. These lakes were originally filled with brackish water and influenced by the Pacific Ocean. However, construction of a water gate in 1963 prevented contact with seawater, and the lakes have subsequently changed from being brackish to containing freshwater. Water flows from Lakes Nishiura and Kitaura into Lake Sotonasakaura, located to the south, before passing through a water gate and entering the Pacific Ocean. Lake Nishiura is the second largest lake in Japan. Since the agricultural activities in the drainage basins of these lakes include the raising of livestock such as pigs, chickens and cattle, and the cultivation of crops, such as rice, lotus root, melons and sweet potatoes, these lakes receive considerable runoff from the agricultural areas. In addition, because the water from Lakes Nishiura and Kitaura is utilized for agricultural, industrial and domestic purposes, the lakes are considered important water sources in this region. The characteristics of the sample sites are shown in Table 1. N1 and K1 (uplake area) are located near the mouths of the Koise and Tomoe rivers, respectively. Water samples from these shallow areas were characterized as having relatively high NO 3 –N concentrations and loss on ignition (LOI) characteristics, especially at K1 (maximum 7000 µg l -1 ), and were markedly influenced by river water supply. Sites N4, N7 and K3 are located near the middle of the lakes where the increased depth results in an accumulation of substances such as particulate organic matter. N8 and K6 are located in the southern part of the lakes (downlake area) where they are drained by the Hitachitone and Wani rivers, respectively. Electrical conductivities (EC) were relatively high and the grain size of bottom sediments was larger than in other lake regions. Field sampling Sampling was performed at monthly intervals in each lake. The sediment–water core samples were collected from four locations in Lake Nishiura (N1, N4, N7 and N8), and Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 166 - from three locations in Lake Kitaura (K1, K3 and K6) from May 2007 to January 2008 (Fig.1). The samples were collected from a boat using a gravity core sampler (converted RIGO, KB Type Core Sampler) and an acrylic column (H500 mm, φ70 mm). Seven samples were collected at each site for estimation of phosphorus release flux from sediments. In addition to the seven samples, samples were also collected for redox potential measurements and nitrate experiments in May, June, and July. Oxidation of these samples was prevented by plugging the samples immediately using rubber stoppers and storing samples in plastic containers for transport to the laboratory. Laboratory experiments Estimation of phosphorus release flux under aerobic/anaerobic conditions Samples collected from each site were analyzed to determine the baseline conditions. The DO concentration and the pH of the water in each acrylic column were measured using a water quality meter (TOA DKK, DO24P, WM22EP). After conducting measurements, the overlying water was removed from the upper portion of the column using silicon tubing and a plastic syringe. The water layer from the surface of the sediment to 10 mm above the surface, defined as the boundary layer, was sampled and distinguished from the water in the water column, or overlying water. The overlying and boundary layer water samples were filtered through membrane filters (Millipore, pore size 0.45 μm) and analyzed for DIP and dissolved total phosphorus (DTP) using a colorimetric method and a potassium peroxodisulfate digestion–colorimetric method (Murphy and Riley, 1962; Menzel and Corwin, 1965) on a continuous flow analyzer (Bran+Luebbe, AACS – II and Auto Analyzer 3), respectively. Dissolved organic phosphorus (DOP) was calculated as the difference between DTP and DIP. Nitrate in the overlying and boundary layer water was measured by the cadmium reduction method (APHA, 2005) using AACS – II. After removing the water, the surface layer of the sediment core samples (0 – -2 cm) was removed and macrofauna were collected and identified from the sediment slices. After weighing each slice, sediment samples were separated from pore water by centrifugation at 3000 rpm for 20 min. DIP in the separated pore water was measured by the same method as DIP in the overlying water. Phosphorus release experiments (14 days) were performed in an incubator in the dark at 25°C using the remaining six samples. Three samples were aerated with ambient air (aerobic conditions), and the other three were covered by rubber stoppers (anaerobic conditions) as shown in Fig.2 (a). The DIP and DTP concentrations in the overlying, boundary layer, and pore water were analyzed by the same methods employed for baseline determination. The DIP release flux was estimated from the difference in DIP concentrations in the overlying and boundary layer water samples measured before and after incubation. Verification of effects of nitrate addition on phosphorus release The effect of nitrate in overlying water on DIP release from the bottom sediments was investigated. The experiments were performed using uplake (N1 and K1) and midlake (N7 and K3) samples collected in July. The uplake samples and one of three midlake samples were used as reference data. The other samples were spiked with two concentrations of NaNO 3 (a1: about 1000 µg N l -1 , a2: 2000 µg N l -1 , respectively) and maintained in an incubator in the dark for 14 days at 25°C under anaerobic conditions (Fig.2 (b)), before DIP release flux was estimated. Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 167 - Data logger Thermo couple Platinum electrode Comparing electrode Sensor spot 0 to 1 cm -1 to -2 cm Non-destructive DO meter Optical fiber Sediment Water Water-sediment column Incubator 25ºC, non-irradiation aerobic condition (air bubbling ) anaerobic condition (N 2 saturation in plastic bag) Plastic bag Incubator 25ºC, non-irradiation Water Sediment Water-sediment column Plastic bag anaerobic condition (N 2 saturation in plastic bag) N1 NO 3 -N N7 K1 K3 K3(a1) K3(a2)N7(a2) NO 3 -N NO 3 -N NO 3 -N 1000ppb 2000ppb 1000ppb 2000ppb N7(a1) Incubator 25ºC, non-irradiation aerobic condition (air bubbling) anaerobic condition (stand alone) Air Whisker Rubber stopper Rubber stopper Water Sediment Water-sediment column Water Sediment Water-sediment column (a) (b) (c) Fig.2 – Schematic diagrams of experimental systems for (a) estimation of phosphorus release flux from bottom sediment, (b) verification of NO 3 on phosphorus release, and (c) measurements of redox potential and DO concentration. Redox potential measurement To assess the oxidation/reduction state in the bottom sediments of each experimental condition (aerobic, anaerobic and nitrate addition) the redox potential (Eh) in the sediments was measured. The measurements were carried out using samples collected in May (anaerobic), June (aerobic) and July (nitrate addition). Eh (mV) measurements was taken by placing the platinum electrode and the comparing electrode (Toa DKK, EL4400) in the sediment (-1 – -2 cm-layer) and was logged every 10 minutes on a data logger (Keyence Japan, NR – 2000). The temperature was measured by Cu – Co thermocouple in the same layer as Eh. During the same period, the DO concentration in the overlying water and sediments (pore water) was monitored using a non-destructive DO meter (PreSens, Fibox 3 – Trace) from outside the acrylic column. Schematic diagrams of the Eh and DO assay systems are shown in Fig.2 (c). Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 168 - RESULTS AND DISCUSSION Seasonal changes in phosphorus concentration and release flux The seasonal changes in boundary-layer and pore-water DIP for each sampling site in Lakes Nishiura and Kitaura are shown in Fig.3 (a) and (b). DIP increased in the boundary layer in the high water temperature period (August – September) at most sample sites in both lakes. Marked increases in DIP were detected at K3 and K6, and the maximum concentrations were 131 and 107 μg l -1 , respectively. DIP increased in both pore water and boundary layer water in the high water temperature period at most sampling sites, as shown in Fig.3 (b). The increases observed in Lake Kitaura were relatively high, and the maximum concentrations during the high water temperature period were 830 and 990 μg l -1 at N7 and N8 in Lake Nishiura, and 1200 and 1560 μg l -1 at K3 and K6 in Lake Kitaura, respectively. Shaw and Prepas (1989) reported that DIP in pore water increased with an increase in water depth. DIP was observed to increase at the mid- and downlake sites in each lake. At the deeper sites in Lakes Nishiura and Kitaura, DIP was also high. DOP in boundary layer and pore water was comparatively lower than DIP. The DOP concentrations were observed to range between 1 – 16 μg l -1 in boundary layer and 0 – 276 μg l -1 in pore water at each site. In addition, defined seasonal changes were not observed. Monthly changes in DIP release fluxes under aerobic and anaerobic conditions and the average DIP flux and standard deviation (n=3) for each condition are shown in Fig.4. There was no seasonal variation in DIP release fluxes, and the fluxes remained low at uplake sites. DIP flux under anaerobic conditions at uplake sites was almost identical to Concentration (µg l -1 ) 0 20 40 60 80 100 120 140 MJ JASONDJ N1 N4 N7 N8 K1 K3 K6 0 400 800 1200 1600 MJ JASONDJ N1 N4 N7 N8 K1 K3 K6 (a) (b) 0 200 400 600 800 1000 M J JA SOND J N1 N4 N7 N8 K1 K3 K6 DIP gradient (µg l -1 cm -1 ) Month (C) Fig.3 – Monthly changes in initial DIP concentrations of (a) boundary layer water and (b) pore water, and (c) DIP gradient across sediment – water interface. Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 169 - DIP release flux (mg m -2 d -1 ) Month 0 5 10 15 20 MJ JASONDJ 0 5 10 15 20 MJ JASONDJ 0 5 10 15 20 MJJA SONDJ 0 5 10 15 20 MJ JASONDJ N1 N4 N7 N8 0 5 10 15 20 MJ JASONDJ 0 5 10 15 20 MJ J A SONDJ 0 5 10 15 20 MJ JASONDJ K1 K3 K6 aerobic anaerobic * * ** ** * * ** * ** * ** * * ** ** ** ** ** * p<0.05 ** p<0.01 ** ** * ** ** * ** ** ** ** * ** ** * * Fig.4 – Monthly changes in DIP release flux from sediment under aerobic and anaerobic conditions. that observed under aerobic conditions. Conversely, at the mid- and downlake sites, DIP fluxes under anaerobic conditions varied seasonally, with the maximum fluxes at each of these sampling sites occurring in August. The DIP fluxes were 3.23, 8.73, 11.25, 15.72 and 18.05 mg m -2 d -1 at N4, N7, N8, K3 and K6, respectively. It is well known that the release of phosphorus from lake sediments depends on water temperature (Marsden, 1989; Søndergaard et al., 1990). Jensen and Andersen (1992) studied the relationship between phosphorus release rates from sediments and three parameters (temperature, nitrate and pH) considered to affect phosphorus release, and found that only temperature was correlated with seasonal variations in sediment phosphorus release rates. Ishii et al. (2007) conducted the DIP release experiment during July 2005 until May 2006 at the water temperature measured at each month, 5 – 25°C, and reported that other factors except water temperature were affecting the spatial variation of DIP release among each sampling points. Since seasonal variations in DIP release flux were verified at most sampling sites, all of the experiments in this study were performed at 25°C. The initial DIP gradient across the sediment–water interface is shown in Fig.3 (c). Gradients increased during the high water temperature period at most sampling sites. Given the increase observed in the DIP concentration of pore water during the high water temperature period, seasonal variation of DIP release flux was considered to proceed from an increase in the DIP gradient. At uplake sites, increases in DIP gradient during the high water temperature period were relatively low. The pH, one of the parameters affecting phosphorus release (Zhou et al., 2005) ranged between 7.0 – 8.7 and 7.0 – 8.0 at N1 and K1, respectively, and was similar to the levels at mid- and downlake sites (7.2 – 8.7). Nitrate concentrations were also high at uplake sites (Table 1). It was reported that high nitrate concentrations at sites near river inflows were likely affected by phosphorus release from sediments (Ishii and Komatsu, 2006; Ishii et al., 2009). The changes in the Eh of sediments under anaerobic conditions are shown in Fig.5. The Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 170 - Eh (mV) -300 -200 -100 0 100 200 300 400 N1 N4 N7 N8 -300 -200 -100 0 100 200 300 400 K1 K3 K6 L Nishiura L. Kitaura 14day 14day Fig. 5 – Changes in Eh under anaerobic conditions for 14-day incubation experiments in sediments from Lakes Nishiura and Kitaura. Eh at uplake sites was relatively higher than at mid- and downlake sites. In particular, at K1, the Eh was positive during the experimental period despite the prevailing anaerobic conditions. At other sites, Eh gradually decreased during the experimental period and was approximately -200 mV by the end of the experiment. DO concentrations in boundary layer water decreased to 0.0 mg l -1 within 30 hours after the start of the experiment, and in sediment (pore water), the concentration was 0.0 mg l -1 for the duration of the experiment. The sediments from the uplake areas were oxidized by nitrate, which also affected DIP release. Contribution of oxygen and nitrate in boundary layer water to Eh in sediment and phosphorus mobilization To examine the contribution of nitrate to DIP release from sediments, incubation of sediment–water core samples with nitrate were carried out under anaerobic conditions and DIP release flux and Eh were measured in the sediment (Fig.2 (b) and (c)). The DO decreased to 0.0 mg l -1 within 30 – 48 hours in boundary layer, and remained at 0.0 mg l -1 in the sediment throughout the experiment. The pH and NO 3 – N values in boundary layer of each sediment–water column are shown in Table 2. The pH was not affected by nitrate addition and showed no change before and after the 14-day incubation. NO 3 – N was not detected at the end of the experiment and the fluctuation in Eh is shown in Fig.6. The addition of nitrate to mid-lake samples (N7 (a1) and K3 (a1); about 1000 µg N l -1 ) resulted in a slight increase in Eh in sediment at N7, but not at K3. Under conditions of high nitrate addition (a2, 2000 µg N l -1 ), Eh increased at both N7 and K3. Eh reached the same levels at both N7 (a2) and N1, which are located in the uplake area, and decreased similarly to that measured at N1. The Eh at K3 (a2) gradually increased before decreasing to the same levels as K3 and K3(a1) after the 14–day incubation. Changes in Eh at N7 and K3 under aerobic conditions were similar to those under anaerobic conditions at each site, despite the boundary layer being saturated with DO as shown in Fig.6. Variation in Eh under aerobic conditions showed similar trends to those under anaerobic conditions at each site (data not shown). Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 171 - Tabl e 2 – pH and NO 3 -N concentrations in overlying water before and after nitrate addition. N1 N7 N7(a1) N7(a2) K1 K3 K3(a1) K3(a2) Initial 7.3 7.4 7.4 7.5 7.4 7.6 7.6 7.6 End 7.5 7.5 7.6 7.6 7.6 7.7 7.7 7.7 Initial 1911 29 773 2145 2100 150 1292 2211 EndNDNDNDND NDNDNDND pH NO 3 -N L. Nishiura L. Kitaura -300 -200 -100 0 100 200 300 400 K1 K3 K3(a1) K3(a2) K3(O2) -300 -200 -100 0 100 200 300 400 N1 N7 N7(a1) N7(a2) N7(O2) Eh (mV) L Nishiura L Kitaura (aerobic) (aerobic) 14day 14day Fig.6 – Changes in Eh under nitrate-addition, aerobic and anaerobic conditions for 14-day incubation experiments in sediments from Lakes Nishiura and Kitaura. ΔEh (mV) 0 100 200 300 N1 N7(a2) K1 K3(a 2) 5 8 11 14day Fig.7 – Comparison of declining tendency of ∆Eh during days 5 – 14 of high nitrate concentration samples. ∆Eh was calculated as the difference between each sample point and N7 (at Lake Nishiura) and K3 (at Lake Kitaura) under anaerobic conditions. Therefore, the DO concentration in boundary layer was only considered to oxidize the sediment surface and to affect DIP release from the sediment. Fig.7 shows a comparison of the decreasing tendency of ∆Eh at N1, N7 (a2), K1 and K3 (a2) during days 5 – 14. The ∆Eh was calculated as the difference between each site and N7 (for Lake Nishiura) and K3 (for Lake Kitaura) under anaerobic conditions. The initial NO 3 – N concentrations at N1, N7 (a2), K1 and K3 (a2) were 1911, 2145, 2100 Journal of Water and Environment Technology, Vol. 7, No. 3, 2009 - 172 - 0 2 4 6 8 10 N1 N7 N7(a1) N7(a2) K1 K3 K3(a1) K3(a2) DIP release flux (mg m -2 d -1 ) Fig.8 – DIP release flux in nitrate addition experiments. and 2211 µg N l -1 , respectively (Table 2), and ∆Eh at N1 and N7 (a2) showed a similar declining tendency during this period. ∆Eh did not vary at K1, but showed a marked decline at K3 (a2). Despite similarities in initial nitrate concentrations, ∆Eh differed between each lake and each sample site. The difference in ∆ Eh was a result of differences in the rate of NO 3 consumption (denitrification and/or oxidized decomposition of organic matter) in sediment. DIP release flux in the nitrate addition experiments is shown in Fig.8. A decrease in DIP release flux in samples from the mid-lake area of Lake Nishiura was observed after the addition of nitrate in this experiment, and the effect of increased nitrate concentration on DIP release was most apparent in boundary layer. At K3 in Lake Kitaura, the slight addition of nitrate was ineffective in suppressing DIP release. It follows that the Eh in sediment would not increase (Fig.6), and that the increase in sediment Eh observed at K3 (a2) DIP release was suppressed by higher nitrate levels, as shown in Fig.8. Relationship between DIP release and nitrate in boundary layer water Laboratory experiments demonstrated that the presence of NO 3 – N in boundary layer water affected DIP release from lake sediments. The relationship between NO 3 – N concentration and DIP release flux under anaerobic conditions is shown in Fig.9 (a). The presence of over 2000 µg N l -1 nitrate in boundary layer severely suppressed DIP release from sediment. Andersen (1982) found that the release of phosphorus in anaerobic hypolimnetic water was inhibited by sufficiently high nitrate concentrations (>1000 µg N l -1 ), which can prevent the release of iron-bound phosphorus (Kleeberg and Kozerski, 1997). At K3 in the middle of the lake, DIP fluxes during autumn and winter (October – January) were significantly lower than the flux observed in May, despite similar DIP gradients (Fig.4). The NO 3 – N concentration in these seasons was relatively high (370 – 980 µg l -1 ) compared to other seasons (30 – 252 µg l -1 ) due to the increasing nitrate concentration in river water, and DIP release was considered to be suppressed by the nitrate supplied from rivers. The relationship between nitrate supply rates and the DIP release flux under anaerobic conditions is shown in Fig.9 (b). Nitrate supply rates from boundary layer to the sediment were estimated based on the assumption that nitrate in the water column is consumed in the sediment during the incubation experiment (14 days). DIP release fluxes declined for nitrate supply rates of approximately 500 µg N m -2 hr -1 . At nitrate supply rates over 1000 µg N m -2 hr -1 , virtually no DIP was released from sediments. These supply rates for mid- and downlake sites might be underestimated, because there was a possibility that the nitrate in water column were exhausted during the first few . N7(a2) K1 K3 K3(a1) K3(a2) Initial 7. 3 7. 4 7. 4 7. 5 7. 4 7. 6 7. 6 7. 6 End 7. 5 7. 5 7. 6 7. 6 7. 6 7. 7 7. 7 7. 7 Initial 1911 29 77 3 2145 2100 150 1292 2211 EndNDNDNDND. (Min.-Max.) pH 7. 7 (7. 0-8 .7) 7. 7 (7. 2-8.0) 7. 8 (7. 2-8.2) 8.1 (7. 4-8 .7) 7. 5 (7. 0-8.0) 7. 8 (7. 4-8.2) 8.1 (7. 7-8.6) EC (µS cm -1 ) 228 (180-266) 286 ( 271 -303) 3 07 (286-355)