The mission of this group is to assess and reduce the environmental risk caused by application of pesticides and other organic chemicals in agro-ecosystems, and also to develop bioremediation techniques to restore environments contaminated with recalcitrant organic chemicals.

The major research topics in FY 2002 were as follows: 1) risk assessment of herbicides in aquatic organisms, 2) simple detection of the endocrine-disrupting effects of organic chemicals (Topic 1), 3) mechanism of induction of systemic acquired resistance to some organic chemicals in plants, 4) molecular genetics and molecular ecology of bacteria that degrade chlorobenzoates and 2,4-D in soil, 5) enhanced photodegradation of dioxins, 6) technology to reduce herbicide outflow from paddy fields into river water, and 7) in situ bioremediation to prevent water pollution derived from herbicide (simazine) application on a golf course (Topic 2).

Nineteen original research papers were published this FY. In September 2002, this group also organized the second Seminar on Organic Chemicals Studies - Current Status of Environmental Pollution Caused by POPs (Persistent Organic Pollutants) - to discuss dynamics in the environment, ecotoxicology, and technology for degradation of POPs.

Topic 1: In vivo evaluation of endocrine-disrupting activities of farm chemicals using Japanese medaka

Over the last few years there has been much concern about the endocrine-disrupting effects of farm chemicals in wildlife. Pesticides applied to paddy fields are apt to flow out to surrounding water bodies such as rivers and lakes, and some pesticides are detected frequently at ppb levels. There is an urgent need to evaluate the endocrine-disrupting effects of these pesticides on aquatic life.

Japanese medaka (Oryzias latipes) has been used for the toxicity assessment of a variety of chemicals because it is easily bred and has a short life cycle. Special attention is currently focused on the d-rR strain of medaka (Photo 1), because a gene controlling body color is located on the Y chromosome. Hence, the sex genotype - either XX (female) or XY (male) - can be easily determined by the body color (white or orange-red, respectively). Further, the presence of sex hormones and their analogs, including estradiol and diethylstilbestrol (DES; an artificial female hormone), determines secondary sexual characters such as the shape of the anal fin and occurrence of oviposition. Therefore, sex-reversal of medaka fish due to hormonal effects can be easily identified by observation of body color and sexual characteristics. We therefore developed a simple testing method to evaluate the endocrine-disrupting effects of farm chemicals in the d-rR medaka strain.

Our method is outlined in Figure 1. Fry of d-rR medaka are exposed for 2 weeks from the day of hatching to the test compound dissolved in 1 L of solution, by using a flow-through system. The test solution is replaced 15 times a day in a semi-automatic manner. After 2 weeks' exposure, the juvenile fish are transferred to fresh water and kept for 40 days until the appearance of secondary sexual characteristics. The effects of chemical exposure are then investigated with respect to weight, length, body color, and secondary sexual characteristics. Finally, the endocrine-disrupting effects are assessed from the sex-reversal ratio.

We used this method to test the endocrine-disrupting activities of 4-nonylphenol (NP). The concentrations tested were 0, 10, 30, 100, and 300 μg/L. No sex-reversal was observed at NP concentrations of 10, 30, and 100 μg/L. When the concentration was increased to 300 μg/L, no sex-reversal was observed in genetic females, but 75% of males showed reversal.

Similarly, we examined the endocrine-disrupting activities of 5 herbicides commonly used in paddy fields: mefenacet, symetryn, benthiocarb, imazosulfuron, and pretilachlor. Although the concentration of each herbicide was set at high levels compared with those commonly detected in the environment, no sex-reversal was detected. With this method, sex-reversed individuals are easily detected by observation of body color and fin shape, without the need for anatomical dissection, and the exposure period is reduced to 2 weeks instead of the 2 months needed with conventional methods. (T. Horio)

 

Topic 2: In situ bioremediation of simazine-contaminated soils on a golf course by using charcoal enriched with consortia of degrading bacteria

In recent years, both scientists and policy makers have shown great interest in technologies for bioremediation of various recalcitrant organic pollutants. We have succeeded in efficiently absorbing and degrading persistent pesticides by mixing bacteria-enriched charcoal with pesticide-contaminated soils under laboratory conditions. To prevent the contamination of subsoils, rivers, and groundwater with the herbicide simazine, which is widely used on golf courses throughout Japan and frequently detected in river water, we conducted a field experiment by laying charcoal enriched with simazine-degrading bacterial consortia (CD7) under the subsoil of a golf course.

On 16 October 2000, we placed a 1-cm-thick layer of charcoal enriched with CD7 consisting of 3 kinds of bacteria under the subsoil at 15 cm deep of a treatment plot in a golf course (Photo 2). In the control plot, charcoal without CD7 was laid in the same manner as in the treated plot. Porous glass cups were inserted at 4 locations in each plot to collect the soil solution directly beneath the charcoal layer. A recommended dose (25 g/a) of water-dispersible simazine powder was applied on 20 October 2000, and after the application we periodically examined changes in the simazine concentration in the soil solution and the soil and charcoal layers, as well as the number of simazine-degrading bacteria in the charcoal. The simazine was applied twice a year (at the end of March and the middle of October) for 2 years. The results were as follows.

1) Simazine concentration in soil solution (Fig. 2): In the control plot, simazine at a concentration of 0.01 μg/L or more was detected at all locations until the seventh week after the first application. In the treated plot, the simazine concentration was 0.005 μg/L or less at all the sampling locations until the sixth week after the first application; it was not detected from the seventh week onward. The simazine-degradation rate in the soil water of the treatment plot reached 92% in 6 months after the first application, compared with the control plot. However, this rate was slightly retarded to 70% after the second application, compared with the control plot. After the third and fourth application, the degradation rate in the soil water of the treated plot was still more than 60%.

2) Number of simazine-degrading bacteria in charcoal (Fig. 2): The initial number of bacteria (7.5 × 10⁷ cfu /g charcoal) had decreased to 106 cfu/g charcoal by 2 months after introduction of the bacteria and had declined to 105 cfu/g charcoal 4 months after introduction. Thereafter, this population level was maintained until the end of the experiment 20 months after introduction.

3) Simazine concentration in the charcoal layer: The concentration of simazine in the charcoal layer was maintained at 5 to 8 mg/kg dry matter until 6 months after the first application in the control plot, owing to the adsorption of simazine. In the treated plot, the concentration reached a maximum (3 mg/kg dry mater) 1 month after the first application, and afterward decreased to 1/20 of the residual amount in the control plot at 5.5 months after the first application. After the second application a similar trend was observed. In the treatment plot, because the simazine absorbed in the charcoal was degraded by the bacteria living within it, the residual amount of herbicide was much less than in the control plot.

Therefore, by laying charcoal enriched with simazine-degrading bacteria consortia under the subsoil of a golf course, we were able to minimize simazine pollution of the subsoil, river water, and groundwater for at least several years. (K. Takagi).

 

The mission of the Heavy Metal Research Group is to elucidate the input-output balance of heavy metals such as cadmium in arable soils and to clarify the mechanisms by which these metals are absorbed and translocated by paddy rice and soybean.

In FY 2002-2003, the group studied the following major topics: 1) role of roots in Cd concentration in soybean seeds (see Highlights in the current report); 2) effects of paddy field water management on the absorption of Cd by rice (Topic 1); 3) evaluation of heavy metal loading of arable soils by irrigation water (Topic 2); 4) application of a 113Cd tracer technique in the evaluation of varietal differences in Cd uptake by soybean and rice; 5) evaluation of heavy metal loading of arable soils by rainfall and fertilizers; 6) bio-availability of Cd in Andosols and Gray Lowland soils; 7) dynamics of Cd in the soybean rhizosphere; 8) screening of rice varieties for low Cd uptake and accumulation in grains; and 9) mechanisms of absorption and translocation of Cd in low-Cd-absorption varieties of soybean and rice.

Dr. S. Ishikawa was given an award by the Japanese Society of Soil Science and Plant Nutrition for his remarkable contribution to the study of aluminum and low-phosphorus tolerance of plants.

 

Topic 1: Appropriate water management to reduce Cd uptake by rice plants

Recently, food safety has become a primary concern of not only consumers, but also producers and agricultural policy makers. The Codex Alimentarius Commission (CAC) created in 1963 by FAO and WHO, is intensively discussing new Cd standards for foods. In light of this we need to minimize the Cd content of staple crops such as rice and soybean.

The amount of bio-available Cd in soils is of primary importance to Cd absorption by rice and thence Cd accumulation in grain. It is a matter of urgency that we develop a promising technology to determine levels of bio-available Cd in soils and formulate effective water management practices for reducing the Cd content of rice grains.

We conducted a pot experiment using 6 soils, collected from different parts of Japan, with different Cd contamination levels. At the relevant rice crop growth stages, soil solution was collected in a porous cup buried 15 cm below the soil surface.

In pots from which the residual water was drained out after heading 7 weeks before harvest, the Cd concentration in the soil solution sharply increased along with a steep increase in the soil reduction-oxidation (redox) potential. Two weeks after heading, the Cd concentration in the soil solution from the drained pots was nearly 100 times as much as that from the pots that were kept flooded until harvest, and the Cd concentration in the soil solution of the drained pots remained high thereafter (Fig. 3, drained pots).

There was an inverse correlation between duration of flooding after heading and Cd concentration in the hulled rice grains. The longer the flooding period after heading, the lower the Cd concentration in the grain (Fig. 4). When soils are flooded, the soil redox potential decreases toward a negative value, and this gives rise to the formation of less-soluble Cd compounds such as CdS, thus reducing the Cd concentration in the soil solution (Fig. 3). It is therefore desirable to keep paddy fields flooded until as close as possible to harvest, unless harvesting works are hindered by the wet soil.

The Cd concentration in the soil solution varies greatly with changes in soil conditions, and, in particular, redox potential. This indicates that the Cd concentration in the soil solution is a good index of bio-available Cd in soils; hence, more promising cultural practices for reducing the Cd content of rice grains could be developed by using the Cd content of the soil solution as an indicator. (Y. Sakurai, K. Sugahara, and T. Makino)

 

Topic 2: Evaluation of Cd loading of arable soils by irrigation

To obtain basic data to estimate the Cd loading of paddy fields by irrigation water, we collected water samples on 2 occasions - in May at 37 points and in August at 26 points - from the Mogami River, which is in one of the main rice production areas in Japan, and we analyzed both the soluble and total concentrations of Cd. The results are shown in Table 1. The soluble Cd concentration in spring (May) samples was lower than that in summer (August) samples, probably owing to dilution by snow melt. River water containing less than 0.05 mg L^-1 of soluble Cd was considered to be unpolluted and was thus considered to contain "background level" concentrations. Water containing more than 0.05 mg L^-1 of soluble Cd was most likely to be polluted either by mining or acidity. However there were no mines upstream of the sampling areas, Cd solubility in some samples appeared to be enhanced by the acidity. The total concentration of Cd was not distinctively different between the spring and the summer samples, probably because of the concentration of suspended solids (SS) in the river water. There was a moderate relationship between total Cd concentration (Cd mg L^-1) and the concentration of SS (mg L^-1) in the background level water. The average total concentration of Cd in this water was 0.116 mg L^-1. It was apparent that water in which the soluble Cd concentration exceeded the background level was either mining-affected or acidic. (T. Saito and S. Murayama)

The mission of the Water Quality and Solute Dynamics Group is to clarify the dynamics of solutes such as nitrate-nitrogen passing through arable lands to water bodies; to develop technologies to monitor loadings of nitrate-nitrogen and other pollutants; and to reduce these loads on the environment. We have 5 ongoing projects: 1) elucidation of the mechanisms of solute movement through the soil and below-ground; 2) development of monitoring methods for nitrate-nitrogen and other pollutants in medium-sized river basins; 3) development of a technology for alleviating the agricultural nitrogen load on the environment by enhancing denitrification; 4) construction of a model adaptable to medium-sized river basins for prediction of nitrogen load and effluent; and 5) examination of the characteristics of dioxin effluents from rice paddy soils to riverine areas. In FY 2002, we successfully estimated detailed water and nitrate fluxes under field conditions (Topic 1). We also developed a new retrieval system for water quality monitoring databases (Topic 2) and a model of SS loadings from arable lands to rivers (Topic 3).

 

Topic 1: Water and nitrate flux through an unconfined aquifer below a cropland

The processes of transport of nitrate leached from crop root zones in arable lands to groundwater and surface water systems are not well understood because of the difficulties of in situ determination of underground fluxes of water and solute. However, it is essential that we have a detailed understanding of field-scale transport of water and nitrate to shallow and deep groundwater bodies if we are to predict changes in nitrate concentrations in aquifers and streams of agricultural watersheds and propose mitigation strategies for non-point-source pollution by nitrate.

We determined the annual horizontal fluxes of water and nitrate through an unconfined aquifer below an Andosol field by numerical analyses based on water-table measurements and nitrate concentration distributions in the field. Equations describing 2-dimensional horizontal groundwater flow, which took into account the spatial distribution of depth to the low-permeability layer, z₀, and the vertical profiles of porosity and saturated hydraulic conductivity in the soil, were solved numerically by a finite difference method.

The material balance of water and nitrate-nitrogen in the subsoil (1~z₀ m) shows that a significant proportion of the water and nitrate in the unconfined aquifer moved vertically downward to the deeper groundwater bodies through the low-permeability layer, which had been recognized as to be impermeable (Fig. 5). The nitrate-nitrogen removal during this period was estimated to be 4.6 kg ha^-1 (Fig. 5). The first-order in situ denitrification rate constant of nitrate, estimated as k = 1.8 × 10^-9 s^-1, was 2 orders of magnitude smaller than those obtained from laboratory experiments using repacked soil columns. These results suggest that denitrification in the unconfined aquifer below the agricultural cropland is a slow process, but that, nevertheless, the travel time of nitrate discharged into the deeper groundwater bodies toward the surface water systems may be large enough for denitrification to become a significant process in the removal of nitrate from watersheds. (S. Eguchi)

Topic 2: System for retrieving information from water quality monitoring databases

Currently, in Japan, various kinds of water monitoring data have been accumulated by local administrative offices and are available to the public. These data have contributed to our understanding of the present status of water quality and also to research on water quality in the environment. For example, if we are constructing a water quality prediction model, we need to verify the accuracy of the model against these long-term monitoring data. However, because such vast numbers of data are not easily handled and analyzed, these databases have not been used to maximum effectiveness by researchers and the public.

Therefore, we developed a PC-based system to facilitate handling of these water quality databases and analysis of data for various purposes. The system enables the user to retrieve water quality data from existing databa-ses. When a user selects a location on the map shown on his or her PC, the water quality and its fluctuations at the location are displayed (Fig. 6). The data can be displayed as tables, pie charts, or bar graphs. The system also can perform various other types of analysis, including multiple regressions, orthogonal polynomials, and analysis of variance. These functions help the user to analyze fluctuations in water quality over years and seasons and the relationships between these fluctuations and other water composition and/or climatic data such as AMeDAS (Automated Meteorological Data Acquisition System) data, which can be retrieved on line. This system is particularly effective for analyzing the relationship between fluctuations in water quality and precipitation. A preliminary analysis with this system clearly showed that fluctuations in water quality were closely related to preceding precipitation. This system is convenient for the utilization and analysis of existing water quality databases. (M. Takeuchi and S. Itahashi)

Topic 3: Model for estimation of suspended solid loadings in "Yatsu" topography

Some herbicides used from the 1960s to the 1980s contained very small amounts of dioxins as impurities. These dioxins were deposited in rice paddy soils during this period, and their efflux to the environment is of public concern. Because dioxins are strongly absorbed by soil particles, the efflux of dioxins occurs together with that of suspended solids (SS) from paddy soils to water bodies. However, it has so far been difficult to estimate how much dioxin is absorbed by SS in this runoff, because there is no effective method of estimating SS loads on a river-basin scale. We therefore developed a model for estimating SS loads in river basins around Ushiku-numa Pond in Ibaraki Prefecture.

These basins consist of rivers and the so-called "Yatsu" topography, in which land-use patterns are typically similar from basin to basin: rice paddies spread along the river at the bottom of the valley, upland fields and residential areas sit on the tops of hills, and forests occupy the slopes in between. For this type of topography, we considered 2 paths of SS flow from various land uses to nearby rivers: one was a direct path from land use to river, and the other was by way of drainage canals in the rice paddy areas to the river. From quantitative analyses of the pattern of land-use chain, we estimated the distributions of the paths of SS flow. Furthermore, from field observations and meteorological data (AMeDAS), we analyzed the relationship between water flow in the rivers and canals and SS concentrations, and incorporated these data into the model. We assumed that some of the SS would settle out on the bed of the canal, so that not all would reach the river. The sedimentation ratio was estimated from the results of a monitoring of SS loadings at both up- and down-stream sites in a canal. This was also incorporated into the model.

We then used the model to estimate the SS loading to a river from various land uses in the river basin (Fig. 7). We estimated that 230 t of SS from 27 km2 of the Inari River basin ran off into the drainage canals between April and December 2002, and about 40% of the SS was trapped by the canals. On the other hand, 34 t of SS ran off from various land uses directly into the river. The total amount of SS reaching the river was estimated at 170 t. This model is effective for estimating the dynamics of SS in a river basin and can be applied to reduction of the SS loading from arable lands to the environment. (S. Itahashi and M. Takeuchi)

The mission of the Dioxin Dynamics Team is to analyze dioxin dynamics in the agro-environment and to develop technologies for remediating dioxin-contaminated soils. In 2002-03, we conducted research in the following areas: 1) dioxin contamination from the environment to staple crops [see Highlights in 2002, 1-1)]; 2) temporal changes in levels of dioxins accumulated in Japanese paddy soils; 3) development of technology for chemical decomposition of dioxins in soils (see Topic); and 4) development of technology for degradation of major dioxins by plant enzymes.

Topic: Chemical decomposition of dioxins in soils

Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) are persistent toxic contaminants. PCDD/Fs are detected at high concentrations in Japanese paddy soils, because some herbicides such as pentachlorophenol (PCP) and 2,4,6-trichlorophenyl 4-nitrophenyl ether (chlornitrofen, CNP), which were used from the early 1960s to the 1990s contained not-inconsequential amounts of PCDD/Fs as impurities. It is imperative that we reduce PCDD/F levels in arable soils to ensure food safety and to minimize the hazardous effects of these chemicals on the aquatic environment. Therefore, we aimed to elucidate the decomposition processes of PCDD/Fs and developed an effective method for chemical decomposition of PCDD/Fs by CaO-rich materials.

Over 99% of octachlorinated dibenzo-p-dioxin (OCDD) in solution was decomposed by this method. During the decomposition process, lower chlorinated congeners of OCDD were not produced, but the production of 3,4,5,6-tetrachloro-1,2-benzenediol was identified by GC-MS scan (Fig. 8). Nearly 80% of radicals (quantified by electron spin resonance, ESR) were produced within 3 min in the decomposition reaction (Fig. 9). These results indicate that PCDD/Fs are decomposed by a radical reaction.

We examined the optimal conditions for degradation of PCDD/Fs in paddy soils (Andosol, Yellow soil, and Gray Lowland soil) (Fig. 10). PCDD/Fs contained in the soils could be decomposed when the proportion of CaO materials constituted 10% or more of the targeted soil (w/w); however, the soil pH increased to more than 12 in all soils immediately and 1 week after addition of the CaO materials remained at this level. Although the decomposition rate of PCDD/Fs in the Andosol was lower than that in the other soils, the soil pH also exceeded 12. Because of this steep increase in soil pH, it is difficult to apply this decomposition method directly to cultivated soils, but it can be applied to uncultivated soils with heavy PCDD/F contamination, such as soils in the vicinity of municipal incinerators. (N. Seike, T. Makino, and T. Otani)