ANALYTICAL SCIENCES JANUARY 2004, VOL. 20 159 2004 © The Japan Society for Analytical Chemistry Determinations of Humic Substances and Other Dissolved Organic Matter and Their Effects on the Increase of COD in Lake Biwa Shinichi AOKI,* Yasuro FUSE,** and Etsu YAMADA*,**† *Department of Chemistry and Material Technology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606–8585, Japan **Center for Environmental Science, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606–8585, Japan Humic substances and other dissolved organic matter (DOM) in Lake Biwa and the surrounding rivers were investigated to elucidate their origins and behavior. An annual increase in chemical oxygen demand (COD) has been observed in the northern basin of Lake Biwa since 1985. The concentrations of dissolved organic carbon (DOC) in the northern and southern basins of Lake Biwa were 1.7 – 2.4 mgC/l and 1.9 – 2.6 mgC/l, respectively. The DOC concentrations tended to be high in summer and low in winter, and the seasonal changes in the concentrations of humic substances were small. The humic substances content of DOM was considered to be comparatively small because the ratio of the concentration of humic substances to DOC was in the range of 0.14 – 0.32. From the results of the fractionation of DOM in lake waters, it was estimated that hydrophobic acids, such as humic substances and hydrophilic acids, were about 25% and 45%, respectively. The main origin of hydrophobic acids in Lake Biwa may be humic substances from soils around the rivers that flow into Lake Biwa, while hydrophilic acids may be due to the inner production by phytoplankton. Therefore, the increase of COD in the northern basin of Lake Biwa may be attributed to the contributions of not only humic substances but also hydrophilic acids. (Received September 25, 2003; Accepted November 5, 2003) hydrous iron oxide (HIO) and their reactivity with Introduction permanganate oxidant. They found that the fraction with adsorptivity to HIO consisted primarily of refractory organic Chemical oxygen demand (COD) and biochemical oxygen carbon.4–6 However, the dynamics of DOC could not be demand (BOD) have been used as the main indexes of water quantitatively explained by these data alone. pollution in lakes and rivers. Recently, total organic carbon In our previous study, humic and fulvic acids in environmental (TOC) and dissolved organic carbon (DOC) are also being used as another indexes of water quality. In Lake Biwa, the largest lake in Japan, the annual increase of COD has been observed since 1985, although the values of BOD and chlorophyll-a have been almost constant. Moreover, the values of COD in the rivers that flow into Lake Biwa have not increased (Fig. 1).1 These results suggest that refractory organic matter may have accumulated in Lake Biwa. To clarify the reason for the increase in COD, it is important to analyze the dynamics of both pedogenic and autochthonous DOC separately. In several lakes, chemical fractionation methods have been applied to analyze the temporal changes of DOC.2 Imai et al. divided DOC occurring in Lake Biwa into six fractions: hydrophobic acid, hydrophobic base, hydrophobic neutral, hydrophilic acid, hydrophilic base, and hydrophilic neutral. As a result, they reported that the hydrophilic acid fraction contained more than 40% of the total DOC, whereas the hydrophobic acid fraction contained less than 30%.3 Hori et al. attempted to characterize organic carbon samples in Lake Biwa and the surrounding rivers according to their adsorptivity to Fig. 1 Annual changes in chemical oxygen demand (COD) and † To whom correspondence should be addressed. biochemical oxygen demand (BOD) in the northern basin of Lake E-mail: [email protected] Biwa.1 , COD; , BOD; , COD–BOD. 160 ANALYTICAL SCIENCES JANUARY 2004, VOL. 20 water were determined by UV spectrophotometry after preconcentration with diethylaminoethyl(DEAE)-cellulose and separation at pH 1 by centrifugation, where the relationship between aquatic humic substances and trihalomethane (THM) formation was further examined.7 Furthermore, simultaneous determinations of the concentration and molecular weights of humic substances in environmental water were developed by gel chromatography with a fluorescence detector. Such methods were applied to the direct determination of humic substances in the Katsura, Uji, Kidzu, and Yodo Rivers.8 In the present study, our method was applied to the determination of humic substances in Lake Biwa and the surrounding rivers to clarify their contributions to the increase of COD in Lake Biwa. Moreover, other dissolved organic matter (DOM) in Lake Biwa and the rivers were investigated to elucidate their origins and behavior. Experimental Reagents and apparatus Humic acid (extracted from peat soil; Aldrich Chemicals) was purified by repeated dissolution in 0.1 M sodium hydroxide and precipitation in 0.1 M hydrochloric acid until the color of the Fig. 2 Sampling stations on Lake Biwa and in the surrounding supernatant solution was eliminated. Fulvic acids obtained rivers. Lake Biwa: L1 Hikone, L2 Biwako-ohashi, L3 Akanoi. from A-horizons of brown forest soil (Dystric Cambisol, Dando, Rivers: R1 Ane River, R2 Amano River, R3 Yagura River, R4 Aichi, Japan) and ando soil (Humic Andozol, Inogashira, Inukami River, R5 Uso River, R6 Yasu River, R7 Katsura River, R8 Shizuoka, Japan) were used. Fulvic acids prepared according to Uji River, R9 Kidzu River, R10 Yodo River. the methods developed by the International Humic Substances Society (IHSS) were supplied by the Japan Humic Substances Society and used as a standard without purification. Procedure for the determination of humic and fulvic acids in Diethylaminoethyl(DEAE)-cellulose (Wako Pure Chemicals, environmental water samples after preconcentration Osaka, Japan) was purified according to the previous study.7 A The environmental water samples were collected from Lake weakly basic anion exchanger with a DEAE group (Toyopak Biwa and the rivers (Fig. 2). The lake water samples were DEAE; Tosoh) was used instead of DEAE-cellulose for the collected at two sites in the northern basin (Hikone and Biwako- direct determination of humic substances. A macroporous resin ohashi) and one site in the southern basin (Akanoi). The river (Amberlite XAD-7HP, 28 – 73 mesh), a strongly acidic cation water samples were collected from the Yasu, Uso, Inukami, exchanger (Bio-Rad AG-MP-50, 100 – 200 mesh), and a Yagura, and Ane Rivers, which flow into Lake Biwa. The river strongly basic anion exchanger (Bio-Rad AG-MP-1M, 50 – 100 water samples were also collected from the Katsura, Uji, Kidzu, mesh) were used for the fractionation of humic substances and and Yodo Rivers.7,8 At the stations, water samples were taken at other DOM. The XAD-7HP resin was soaked in 0.1 M sodium the depth of 0.5 m. All water samples were filtered through a hydroxide, washed with pure water, and sequentially purified by membrane filter (0.45 µm, Millipore) as soon as possible to Soxlet extraction with methanol, diethyleter, acetonitrile, and avoid biodegradation. The membrane filters were used after methanol. The ionic exchangers were washed by Soxlet extraction rinsing with 1 M hydrochloric acid and distilled water. with methanol and further washed with pure water. All other Preconcentration and determination of humic and fulvic acids chemicals were of the best commercial grade. Pure water was in river water samples were carried out according to the prepared by a Millipore Milli-Q water purification system. methods used in the previous study.7 The apparatus used for HPLC was a Shimadzu LC-10AD chromatography pump equipped with a Shimadzu RF-535 Procedure for the direct determination of humic substances in fluorescence detector. The excitation and emission wavelengths environmental water samples8 were 340 nm and 435 nm, respectively. A Shimadzu C-R7A A 100 µl aliquot of filtrated water samples without chromatopac was used for data analysis. The test samples were preconcentration after adjusting to 0.01 M sodium hydroxide applied to a gel filtration column, Superose12 HR10/30 (300 × solution was measured by gel chromatography with a 10 mm i.d.); Pharmacia Biotech, using a sample injector fluorescence detector (chromatogram A). Further, a filtered (Rheodyne 7125) with a 100 µl loop. A 0.01 M sodium water sample passed through an anion exchanger with DEAE was hydroxide solution was used as an eluent at a flow rate of 0.4 ml measured similarly (chromatogram B). The concentration and min–1. Two TOC meters, a Shimadzu TOC-5000 and a Uniflex molecular weights of humic substances in river water samples TOC-1010, were used for the determination of DOC. A were calculated by means of the differential chromatogram (C) Shimadzu UV-160A spectrophotometer was used for the obtained as the difference of the chromatograms (A) and (B). measurement of UV-absorbame (260 nm) of water samples. A The concentration of humic substances was calculated with Kubota KN-70 centrifuge was used for the separation of humic reference to the standard of fulvic acid because fulvic acid is substances. A Horiba F-15 pH meter, a Denki-Kagaku Keiki predominant in environmental waters. AOL-40 EC meter, and a Mitsubishi-Kagaku TN100 TN meter were used for the measurement of pH, electric conductivity, and Procedure for the fractionation of DOM in environmental water trace total nitrogen in environmental water samples, samples respectively. The fractionations of DOM in Lake Biwa and the rivers were ANALYTICAL SCIENCES JANUARY 2004, VOL. 20 161 Table 1 Analytical results for humic substances (HS) and dissolved organic carbon (DOC) in Lake Biwa and the surrounding rivers HS DOC HS/DOC Sampling stationa Dateb mg l–1 MeanmgC l–1 Mean Mean Lake Biwa Hikone (L1) A 0.44 2.8 0.16 B 0.39 1.8 0.22 C 0.33 1.8 0.18 Biwako-ohashi (L2) A 0.18 1.9 0.09 B 0.34 1.7 0.20 C 0.31 1.9 0.16 Akanoi (L3) A 1.13 3.0 0.38 B 0.60 1.9 0.32 C 1.27 2.5 0.51 Rivers flowing into Lake Biwa (6 rivers) (R1 – R6) A 0.28 – 1.04 0.61 0.7 – 2.2 1.4 0.28 – 0.53 0.42 B 0.34 – 1.09 0.56 1.0 – 1.7 1.5 0.28 – 0.64 0.38 C 0.23 – 1.76 0.66 0.6 – 3.1 1.3 0.39 – 0.62 0.48 Yodo rivers (4 rivers) (R7 – R10) B 0.50 – 1.53 0.96 1.9 – 2.7 2.2 0.26 – 0.61 0.43 a.