Leaching and degradation of triasulfuron, metsulfuron-methyl, and chlorsulfuron in alkaline soil profiles under field conditions.
Triasulfuron, metsulfuron-methyl, and chlorsulfuron are the 3 most commonly used sulfonylurea herbicides for controlling broad-leaved weeds in the cereal crops of southern Australia. The compounds are weak acids (p[K.sub.a] = 3.3-4.5) and solubilities in water, which vary with pH, range from 300 to 28000 mg/L (Beyer et al. 1988; Iwanzik and Amrein 1988). The herbicides are applied at very low application rates ranging from 7 to 30 g/ha. They possess unprecedented herbicidal activity, and susceptible crops such as legumes and pastures are highly sensitive to trace level residues of the herbicides in the soil (Brown 1990; Moyer et al. 1990). Therefore, the persistence of these herbicides could impact on the performance of subsequent crops.
The degradation pathways of these herbicides include chemical hydrolysis and microbial degradation. However, in alkaline soils, hydrolysis of sulfonylureas is very slow (Sabadie 1996; Sarmah et al. 1998), leaving biodegradation as the dominant pathway that determines their persistence in such soils (Ravelli et al. 1997; Sarmah et al. 1999).
These herbicides exist in anionic form in agricultural soils with pH [is greater than] 6.0 and are weakly adsorbed (Walker et al. 1989; Sarmah et al. 1999). Therefore, they have high potential to leach under conditions of high rainfall and consequently may reach the zones of low organic matter and little biological activity. Due to the slower rate of degradation, leaching can result in their longer persistence in the subsoils and may increase the risk of phytotoxicity to succeeding crops. For instance, from a laboratory incubation study, Sarmah et al. (1999) recently reported that the rate of chlorsulfuron degradation was 4 times slower at 30-40 cm depth than in the topsoil (0-10 cm). The effect of reduced organic matter and low biological activity in the deeper layers of the soil was clearly reflected in their study. The pH of the soils used in the study ranged from 8.7 in the 0-10 cm layer to 9.4 at 30-40 cm depth. Since microbial activity in soil is a function of several variables (e.g. temperature, water content, and organic matter content), the persistence of herbicides in the field will be influenced by the combination of these factors operating simultaneously.
A recent review (Sarmah et al. 1998) on the fate and behaviour of sulfonylurea herbicides revealed that most research work under field conditions has been carried out in soils with pH values ranging from acidic to slightly alkaline (pH [is less than] 8.0). In such soils, the herbicides can move to significant depth in the soil profile (Walker and Welch 1989; Vicari et al. 1994; Stork 1995; Walker and Robinson 1996). However, little is known about the persistence and leaching of sulfonylureas in highly alkaline soils (pH [is greater than] 8.0). A study on such soils of southern Australia was reported by Stork (1995), in which chlorsulfuron moved to a depth of 50 cm in a gradationally textured alkaline loam soil (pH 8.5-9.5 in the 0-50 cm depth) more than 3 months after initial application, at which time the cumulative rainfall was 103 mm. In contrast, Walker and Robinson (1996) found no evidence of chlorsulfuron residues persisting at 4 alkaline sites (pH 7.6-8.8) in southern Queensland after 14 years of repeated application. Temperature may have been a significant factor resulting in rapid degradation of chlorsulfuron under tropical conditions in Queensland. It is important to note that all of these studies involved herbicide analysis by bioassay, which makes comparison between studies difficult due to the use of different plant species.
The purpose of this study was to assess the persistence of triasulfuron, metsulfuronmethyl, and chlorsulfuron herbicides under field conditions and to obtain a more realistic assessment of their behaviour in highly alkaline soil profiles by using a newly developed chemical assay. The specific objectives were to study the leaching behaviour of a non-reactive tracer ([Br.sup.-]), triasulfuron, metsulfuron-methyl, and chlorsulfuron in alkaline soil under 2 different rainfall regimes, and to investigate the dissipation mechanisms from the soil profile.
Materials and methods
The experimental site was located on a calcareous sandy loam soil near the township of Avon (34 [degrees] 17'S and 138 [degrees] 20'E) in South Australia, with an average annual rainfall of 320 mm. The site is typical of the cereal cropping region in southern Australia where rainfall is dominant in winter. Selected physical and chemical properties of the soil are presented in Table 1. The site had no previous history of sulfonylurea herbicide application.
Table 1. Physical and chemical properties of the Avon sandy loam soil
Depth pH Clay Sand Silt (cm) (1:5 [H.sub.2]O) (%) (%) (%) 0-10 8.7 12.5 73.3 14.2 10-20 8.8 16.6 69.4 14.0 20-30 9.1 13.5 74.3 12.2 30-40 9.4 17.7 72.3 10.0 40-50 9.7 5.2 69.4 25.4 50-60 9.8 1.0 65.4 33.6 60-70 9.9 0.0 53.6 46.4 70-80 9.8 0.0 53.6 46.4 80-90 10.0 0.0 16.0 84.0 90-100 10.1 0.0 15.9 83.1 Depth OC Microbial C (cm) (%) ([micro]g/g) 0-10 1.6 818 10-20 0.95 500 20-30 0.74 427 30-40 0.44 383 40-50 0.23 -- 50-60 0.17 -- 60-70 0.11 -- 70-80 0.10 -- 80-90 0.09 -- 90-100 0.09 --
The experiment had a completely randomised block design. There were 2 treatments, `rainfall' and `rainfall + irrigation', each applied to 3 replicate plots. Triasulfuron, metsulfuron-methyl, and chlorsulfuron were applied (150 g a.i./ha) to all plots, along with a non-reactive tracer (potassium bromide, KBr) to follow the movement of water in the soil. The plots were 4 m by 1 m, with a 1-m buffer zone around each plot.
Weather data, such as daily rainfall and maximum and minimum soil temperatures, were recorded at a weather station in close proximity to the experimental site (Fig. 1). The study period was chosen to follow the cereal growing season, but the plots were left bare.
[Figure 1 ILLUSTRATION OMITTED]
Herbicide and bromide application
Commercial grade samples of Logran[R] (714 g a.i./kg), Ally[R] (600 g a.i./kg), and Glean[R] (750 g a.i./kg), triasulfuron, metsulfuron-methyl, and chlorsulfuron, respectively, were applied at 150 g a.i./ha, and a non-reactive tracer (KBr) was applied at 50 kg/ha. The amounts of the 3 herbicides and KBr required for each plot were mixed thoroughly in 10 L of water and the mixed solutions were applied on 6 June 1997 to all the plots by using a hand-held sprayer, which had a shrouded nozzle to minimise drift across the plots. The treatments were immediately incorporated into the soil to a depth of about 10 cm with a hand-hoe. The nominal concentrations of herbicides and bromide in the 0-10 cm layer of soil after spraying and incorporating were 100 [micro]g/kg and 33 mg/kg, respectively, calculated from the measured mean bulk density of 1500 kg/[m.sup.3].
A rainfall simulator was used to apply a calibrated amount of irrigated water ([is greater than] 90% application uniformity) at 6.25 mm/min over 3 plots on selected days, while the other 3 plots received only rainfall. The high rates of irrigation application were chosen to see if preferential flow would occur in these soils. The rainfall simulator was a rotating disc type developed by Malinda et al. (`992). Fig. 1 shows the day and the amount of irrigation applied, together with the sampling days during the experiment.
Soil samples were collected from 0-10 cm depth on the first day, approximately 30 min after applying the herbicide and bromide solution, to determine initial concentrations of the chemicals and soil water content. To avoid cross-contamination, intact soil cores (50 mm diameter) were obtained for all soil samplings. For this purpose, a split-core metal soil sampling tube connected to a trailer-mounted hydraulic ram was used. The hydraulic ram pushed the split tube to the desired sampling depth and cores were extracted to a depth of 60 cm in the profile 19, 28, 46, and 69 days following application. From each plot, 4 cores were extracted and sliced at 10-cm intervals and placed in plastic bags and sealed. The tube was cleaned with a brush and washed with water before the next core was extracted. The last sampling was done to a depth of 160 cm. The holes created by the sampling tube were filled with soil from the nearby buffer zones.
Sample extraction and analysis
For each depth, the 4 soil core samples taken from each plot at each sampling time were bulked and their water contents were determined gravimetrically. The rest of the soil was air-dried, ground, sieved (2 mm), and stored at 4 [degrees] C until analysis. The herbicides from the samples were extracted using a solid-phase extraction technique followed by HPLC and UV detection. A detailed description of the technique can be found elsewhere (Sarmah and Kookana 1999). In brief, samples containing the herbicides were extracted with de-ionised water and passed through a glass micro-fibre filter. The pH of the solution was adjusted down to 2.5. Solid phase extraction was performed using [C.sub.18] extract-clean cartridges, and the residues retained on the cartridges were eluted with acidified methanol and then immediately analysed by HPLC. The detection limits for the herbicides were 1 [micro]g/L in water and 3 [micro]g/kg in soil. For the analysis of bromide in the soil samples, approximately 10 g soil was placed in a 50-mL centrifuge tube to which 20 mL of de-ionised water was added. The tube was shaken for I h and centrifuged for 10 min at 2000 r.p.m. and the supernatant was decanted through a 0.45-[micro]m filter (Minisart). The concentration of bromide was determined using ion chromatography (Dionex Corp, Sunnyvale, CA). The method used a Dionex AS4A-SC column with an alkanol quaternary ammonium functional group, 13-[micro]m-diameter particles with 160-nm-diameter latex beads with 0.5% cross-linking. Eluent was a mixture of 1.8 mM [Na.sub.2][CO.sub.3] and 1.7 mM Na[HCO.sub.3].
Results and discussion
Soil water content
The 1-dimensional profiles of the volumetric water content under the 2 water treatments are shown in Fig. 2. The standard errors of the mean at each depth are relatively small for field plots, and indicate close agreement between replicates. The soil water content was clearly affected by the amount of rainfall/irrigation the plots received. For example, the water content of the soil that received only rainfall ranged from as low as 0.12-0.16 [m.sup.3]/[m.sup.3] in the 10-20 cm depth and 0.11-0.13 [m.sup.3]/[m.sup.3] in the 50-60 cm depths. However, when rainfall was supplemented with irrigation, significant increases in water contents were observed, with the maximum being 0.30 [m.sup.3]/[m.sup.3] in the 10-20 cm depth after a cumulative rainfall of 158.6 mm (Fig. 2). A peak at 20 cm depth was found in both treatments, representing a sharper wetting front, and a second peak was observed at 100-120 cm depth, after a cumulative rainfall and irrigation of 158.6 mm. This bimodal distribution of water in the soil profile indicated that preferential flow was occurring.
[Figure 2 ILLUSTRATION OMITTED]
Results of bromide leaching on different occasions under the 2 water treatments are presented in Fig. 3. Bimodal distribution of [Br.sup.-] was apparent in the rainfall + irrigation treatment compared with rainfall alone. On most occasions, the largest concentrations of bromide were detected in the 10-20 cm depth for both treatments, and after 31.6 mm of cumulative rainfall, most of the bromide still remained between 10 and 20 cm depth. A small peak may have been present in the 50-60 cm depth, which received only rainfall, but there was a much greater peak at 100-110 cm depth after 158 mm of rainfall + irrigation (Fig. 3). Some [Br.sup.-] moved to a depth of 150 cm and possibly beyond.
[Figure 3 ILLUSTRATION OMITTED]
The spatially irregular distribution of bromide indicates that there was some preferential movement via cracks or similar structural elements in soil. The pattern of bromide distribution under the treatment rainfall + irrigation could be linked to the presence of `mobile' and `immobile' water (White 1985) within the soil matrix. When bromide was initially applied to the dry soil, the capillarity of the dry soil would pull the bromide solution into small pores (the `immobile' region) within the aggregates in the topsoil. During subsequent rainfall or irrigation, flow would be through the larger pores in `mobile water', mostly bypassing the small pores containing the bromide, making it relatively immune to leaching (Tillman et al. 1991). However, the bromide subsequently diffused out of the `immobile' region to the `mobile' region where it was more prone to leaching by subsequent rainfall events.
Mass balance for bromide
The mass recoveries (%) of bromide for each water treatment are presented in Tables 2 and 3, in terms of the chemical recovered for the total depth sampled. In the plots which received only rainfall, 80-121% of the applied mass of bromide was recovered, while under rainfall + irrigation, recoveries ranged from 64% to 131%. Within any treatment, however, recoveries of bromide were greater in the early part of the experiment when rainfall was not sufficient to leach the bromide deep into the soil profile. Unusually higher recoveries of [Br.sup.-] observed under both treatments at the beginning of the study period are attributed to large total numbers of cores and the within-plot variability.
Table 2. Average percent mass recoveries ([+ or -] s.e.) of bromide, triasulfuron, metsulfuron-methyl, and chlorsulfuron from soil under different amounts of rainfall
Rainfall Day (mm) Bromide Triasulfuron 19 17.7 121.1 ([+ or -] 10.32) 86.0 ([+ or -] 3.93) 28 18.2 94.5 ([+ or -] 10.29) 35.4 ([+ or -] 3.66) 46 31.6 80.3 ([+ or -] 12.01) 32.4 ([+ or -] 2.59) 69 69.1 89.5 ([+ or -] 3.37) 34.5 ([+ or -] 9.94) Rainfall Day (mm) Metsulfuron-methyl Chlorsulfuron 19 17.7 87.4 ([+ or -] 13.20) 55.8 ([+ or -] 1.19) 28 18.2 68.4 ([+ or -] 16.67) 40.2 ([+ or -] 6.19) 46 31.6 38.8 ([+ or -] 4.11) 29.5 ([+ or -] 4.81) 69 69.1 52.7 ([+ or -] 10.60) 29.3 ([+ or -] 2.93)
Table 3. Average percent mass recoveries ([+ or -] s.e.) of bromide, triasulfuron, metsulfuron-methyl, and chlorsulfuron from soil after different amounts of rainfall + irrigation
Rainfall Day +irr. (mm) Bromide Triasulfuron 19 77.7 131.6 ([+ or -] 20.2) 64.0 ([+ or -] 6.03) 28 90.0 64.4 ([+ or -] 3.53) 44.3 ([+ or -] 6.60) 46 121.1 64.1 ([+ or -] 7.45) 39.4 ([+ or -] 2.09) 69 158.6 102.7 ([+ or -] 5.79) 35.2 ([+ or -] 1.14) Rainfall Day +irr. (mm) Metsulfuron-methyl Chlorsulfuron 19 77.7 64.4 ([+ or -] 7.01) 45.5 ([+ or -] 3.96) 28 90.0 69.2 ([+ or -] 12.92) 43.8 ([+ or -] 1.84) 46 121.1 49.9 ([+ or -] 2.74) 38.2 ([+ or -] 2.88) 69 158.6 52.2 ([+ or -] 7.22) 40.5 ([+ or -] 1.44)
From a solute transport study during intermittent water flow in a field soil, Tillman et al. (1991) reported the average recovery of [Br.sup.-] to vary from 101% to 112%. Considering that [Br.sup.-] is essentially a non-reactive, non-adsorbed, and non-degradable tracer, good recoveries are expected. However, low recoveries of [Br.sup.-] (about 64%) were also observed in this study on sampling Days 28 and 46 under those plots that received rainfall as well as irrigation. This may have been due to macropore flow (Starr and Glotfelty 1990; Steenhuis et al. 1990) carrying some of the [Br.sup.-] below the depth of sampling. In an experiment on a sandy soil, Flury et al. (1995) found the mass recovery to range from 80% to 100% in some plots, but only from 50% to 60% in others. The authors attributed their low recoveries to preferential flow through cracks. In the present study, during the first 3 sampling events, it was not possible to extract cores beyond 60 cm, and therefore, any [Br.sup.-] which had moved beyond this point was not accounted for in the mass balance calculation, which perhaps explains the lower recoveries of [Br.sup.-] (64%) obtained on Days 28 and 46 under rainfall + irrigation (Table 3). During the same period, plots receiving only rainfall showed significantly greater recoveries (80-94%).
Leaching and degradation of the herbicides
Figs 4 and 5 show the distributions of the 3 herbicides in the soil profile. The herbicides showed greater mobility in the plots that received both rainfall and irrigation than those plots which received only rainfall. For example, 69 days after treatment, movement of triasulfuron was restricted to only 40 cm depth after a cumulative rainfall of 69 mm (Fig. 4). In contrast, when rainfall was supplemented with irrigation, triasulfuron reached a depth of 70 cm in the profile (Fig. 5). Of the 3 compounds, metsulfuron-methyl showed the greatest mobility, reaching 120 cm depth after a cumulative rainfall of 158.6 mm (rainfall + irrigation), followed by chlorsulfuron and triasulfuron, which reached 80 and 70 cm depths, respectively (Fig. 5). The relative order of mobility of the herbicides observed in this study, namely, metsulfuron-methyl [is greater than] chlorsulfuron [is greater than] triasulfuron, was similar to that obtained from leaching studies on intact cores of a New Zealand soil (Rahman and James 1989) and with a calcareous soil from southern Australia (Sarmah 1998). It is noteworthy that the order of mobility obtained for these compounds in the present field study is consistent with the corresponding [K.sub.d] values of the compounds, metsulfuron-methyl (0.1 mL/g), chlorsulfuron (0.11 mL/g), triasulfuron (0.26 mL/g), and their dissociation constants (Sarmah 1998). The lower the dissociation constant, the lower the [K.sub.d] value and, hence, the greater the mobility of the compound in these soils. Similar results were also reported by Walker and Welch (1989) in another field study (pH 6.5-7.0) conducted in the UK.
[Figures 4-5 ILLUSTRATION OMITTED]
Although a number of studies have been reported on the leaching and persistence of these herbicides in soils ranging from acidic to slightly alkaline (e.g. Vicari et al. 1991, 1994; Stork 1995), comparison between studies is difficult due to differences in the approaches to analysing herbicide residues in soil. The use of different plant species (Moyer et al. 1990; Kotoula-Syka et al. 1993; Vicari et al. 1994; Stork 1995) in bioassay techniques makes the herbicide analysis plant-species dependent, which can lead to significant variations. For example, Vicari et al. (1994), using a nasturtium bioassay, reported that chlorsulfuron residues moved to 30-50 cm depths at 4 alkaline sites (pH 7.8-8.2) in Italy. In Australia, Stork (1995), using a lupin bioassay, reported 46% of the applied chlorsulfuron leached below 50 cm in a gradationally textured alkaline sandy loam soil (pH 8.5-9.5). Although there were differences in the analytical techniques in these studies, all showed that sulfonylureas moved to significant depths, and their greater mobility in alkaline soils than in acidic or neutral soils was apparent.
Recoveries of the 3 herbicides varied significantly in each of the treatments (Table 2 and 3). In general, metsulfuron-methyl showed slightly greater recoveries (52-87% under rainfall, and 52-64% under rainfall + irrigation) than chlorsulfuron or triasulfuron. Initial low recoveries of chlorsulfuron compared with the other compounds on the first sampling date (after 17.7 and 77.7 mm rainfall) can be attributed to more rapid degradation of the herbicide by the microorganisms (Joshi et al. 1985). Consistent with water content and [Br.sup.-], bimodal distributions of the 3 herbicides were also observed under the treatment of rainfall + irrigation (Fig. 5). The peaks which were observed for metsulfuron-methyl and chlorsulfuron in the 70-80 cm depth and for triasulfuron in the 50-60 cm depth, after a cumulative rainfall of 158.6 mm, were attributed to preferential flow within the individual plots, possibly as a result of the irrigation method used in this study.
Preferential flow has been reported earlier in field studies by Ghodrati and Jury (1992) for atrazine, napropamide, and prometryn on a loamy sand soil, and by Trojano et al. (1993) for atrazine. For instance, Ghodrati and Jury (1992) concluded from their study that peaks of the herbicide residues observed at some depths deeper in the profile were the result of the irrigation methods (intermittent sprinkling, continuous sprinkling, ponding) used in their study. Intermittent heavy rain is more likely to give rise to macropore flow than a slow steady trickle irrigation. In the present study, we employed a rainfall simulator to generate artificial rainfall and produced the pattern of rainfall similar to that of intermittent or continuous sprinkling. Thus, the results obtained here are not surprising. Furthermore, sandy soils are known to be susceptible to leaching of pesticides (Flury et al. 1995) and the phenomenon of preferential flow can be expected to occur to a much greater degree in sandy soils, which have large differences in hydraulic conductivity between the wet and dry states, than in fine-textured soils (Kung 1990a, 1990b).
Degradation of the 3 compounds was rapid during the first 28 days after application, although some delayed degradation was also observed for triasulfuron and metsulfuronmethyl. However, in general, degradation proceeded at a much slower rate thereafter. The initial rapid degradation of the herbicides can be attributed to the substantial biological activity in the topsoil where the herbicides were available to microbes before the first major rainfall. An examination of total microbial biomass (Table 1) in the soil profile revealed that the topsoil contained more than twice the biomass of soil at 30-40 cm depth. As time elapsed, the herbicides were leached into the subsoil where the rate of degradation slowed because of reduced biological activity and higher pH. Initially more than 65% of the applied chlorsulfuron was lost in 28 days, while approximately 50% of triasulfuron and metsulfuron-methyl was lost during the same period (Fig. 6). The times for 50% loss of triasulfuron, metsulfuron-methyl, and chlorsulfuron (18, 30, and 28 days, respectively, under rainfall treatment) were essentially similar to those under rainfall + irrigation treatment (20, 35, and 32 days, respectively). During the first few sampling days, cores were extracted only down to a depth of 60 cm. It is possible that in the treatment rainfall + irrigation, some herbicides were lost beyond the depth sampled. Therefore, the shallow sampling depth may have resulted in an overestimation of this loss. It can be observed that under the treatment rainfall + irrigation, there was very little difference in the rate of degradation between the 3 compounds. However, chlorsulfuron degradation was found to be somewhat faster than triasulfuron and metsulfuron-methyl under the rainfall only treatment (Fig. 6).
[Figure 6 ILLUSTRATION OMITTED]
Total residual mass of triasulfuron, metsulfuron-methyl, and chlorsulfuron remaining in the soil as a function of time is plotted in Fig. 6. A linear regression analysis of the natural logarithm of herbicide concentration in the soil against sampling time gave reasonably good correlation coefficients, but the data tended to support an initial rapid rate of degradation followed by a slower rate. The estimated first-order half-lives of the herbicides varied from 32 to 45 days under rainfall and 46 to 59 days under rainfall + irrigation. While adherence to first-order kinetics under field conditions has also been reported by other workers for chlorsulfuron (Vischetti and Businelli 1992; Vicari et al. 1994), deviations have been observed for metsulfuron-methyl (James et al. 1995).
Based on the half-lives obtained in the present study, the rate of degradation for the herbicides followed the order chlorsulfuron [is less than] triasulfuron [is less than] metsulfuron-methyl under the treatment of rainfall, and triasulfuron [is less than] chlorsulfuron [is less than] metsulfuron-methyl under rainfall + irrigation. A direct comparison of these values of half-lives for the compounds against the reported values in the literature is difficult, because of the limited number of field experiments reported and the considerable variation in the data obtained. This can be expected, given the differences in soil and environmental factors such as soil pH, organic matter content, water content, and temperature in Australia and overseas (Sarmah et al. 1998). For instance, from a field study in Italy, Vicari et al. (1994) reported the half-lives of chlorsulfuron to vary from 51 days to as high as 149 days in 4 locations (pH 7.8-8.2). In contrast, Strek (1998) found the half-lives of chlorsulfuron to be only 18 days in an irrigated Californian soil (pH 6.3-6.9 in the 0-90 cm depth), presumably because of an acidic pH range and a high soil temperature at the site.
It is clear from the foregoing discussion that the sulfonylurea herbicides are highly mobile and their rates of degradation are dependent on the interplay of several soil and environmental factors. The study demonstrated that when rainfall was low and before the addition of supplemental irrigation, most of the herbicides remained in the surface layers, undergoing a rapid phase of breakdown. As time elapsed, the herbicides were leached into the soil of low biological activity and higher pH. This resulted in slower degradation due to lower biological activity and organic matter in the subsoil. Despite a subsequent higher input of water (rainfall + irrigation), the rates of loss under the two water treatments were similar. It is possible that in the rainfall + irrigation treatment, some herbicides may have been lost beyond the depth sampled. This would have resulted in overestimation of loss during the first 3 samplings. The study thus highlights the importance of rainfall in leaching of these herbicides in highly alkaline soils (Vicari et al. 1994), and their rate of degradation is dependent on the time and the amount of precipitation the site receives, as well as the prevailing soil and other climatic conditions.
The results of this study show the high mobility of these herbicides in alkaline soils leading to leaching to the subsoil early in the season and consequently a slower rate of breakdown. Bimodal distributions of water content, [Br.sup.-], and the herbicides suggest the presence of preferential flow when there is excessive moisture. Therefore, mobility can be greater through preferential flow when major leaching events (high rainfall) follow immediately after application of the herbicides into the soil. The results obtained clearly demonstrate that under conditions conducive to leaching (such as in high rainfall periods), the herbicides are likely to reach the zones in the profile with reduced organic matter and biological activity which may result in their prolonged persistence in these soils. The order of mobility for the herbicides was found to be metsulfuron-methyl [is greater than] chlorsulfuron [is greater than] triasulfuron under both treatments. The estimated first-order half-lives varied from 32 to 45 days under the treatment of rainfall, while they were marginally higher (45-59 days) under the treatment rainfall + irrigation. It is envisaged that if the amount of rainfall is not sufficient to leach the herbicide down into the soil profile, and a greater proportion of the applied herbicide residue remains in the top layers of soil, the breakdown could be faster.
The financial support of the Grains Research and Development Corporation (GRDC) of Australia is gratefully acknowledged. We thank Andrew Kerekes (CSIRO Land and Water, Adelaide), and Noel Pederson and Shaun Seigert (South Australian Research and Development Institute, Adelaide) for their help in soil core collection.
Beyer EM, Duffy MJ, Hay JV, Schlueter DD (1988) Sulfonylureas. In `Herbicides: chemistry, degradation and mode of action'. (Eds PC Kearney and DD Kaufman) pp. 117-189. (Marcel Dekker: New York)
Brown HM (1990) Mode of action, crop selectivity and soil relations of the sulfonylurea herbicides. Pesticide Science 29, 263-281.
Flury M, Leuenberger J, Studer B, Fluhler H (1995). Transport of anions and herbicides in a loamy and a sandy field soil. Water Resources Research 31, 823-835.
Ghodrati M, Jury WA (1992) A field study of the effects of soil structure and irrigation methods on preferential flow of pesticides in unsaturated soil. Journal of Contaminant Hydrology 11, 101 125.
Iwanzik W, Amrein J (1988) Triasulfuron behaviour in soil. In `Proceedings 1988 EWRS Symposium: Factors affecting herbicidal activity and selectivity.' pp. 307-312. (Ponsen and Looigen: Wageningen, The Netherlands.)
James TK, Klaffenbach P, Holland P, Rahman A (1995) Degradation of primisulfuron-methyl and metsulfuron-methyl in soil. Weed Research 35, 113-120.
Joshi MM, Brown HM, Romesser JA (1985) Degradation of chlorsulfuron by soil microorganisms. Weed Science 33, 888-893.
Kotoula-Syka E, Eleftherohorinos IG, Gagianaa AA, Sficas AG (1993) Phytotoxicity and persistence of chlorsulfuron, metsulfuron-methyl, triasulfuron and tribenuron-methyl in three soils. Weed Research 33, 355 367.
Kung K-JS (1990a). Preferential flow in a sandy vadose zone, 1. Field observation. Geoderma 46, 51-58.
Kung K-JS (1990b). Preferential flow in a sandy vadose zone, 2. Mechanism and implications. Geoderma 46, 59-71.
Malinda DK, Dubois BM, Pederson RN, Darling WR, Fawcett RG (1992) Improved equipment and methods for the study of rainfall infiltration, run-off, soil and nutrient losses. Department of Agriculture, South Australia, Technical Paper no. 33, pp. 1-26.
Moyer JR, Esau R, Kozub GC (1990) Chlorsulfuron persistence and response of nine rotational crops in alkaline soils of southern Alberta. Weed Technology 4, 543-548.
Rahman A, James TK (1989) Comparative mobility of nine sulfonylurea herbicides in soil column. In `Proceedings: 12th Asian-Pacific Weed Science Society'. Seoul, Korea. pp. 213-217.
Ravelli A, Pantani O, Calamai L, Fusi P (1997) Rates of chlorsulfuron degradation in three Brazilian oxisols. Weed Research 37, 51-59.
Sabadie J (1996) Alcoholysis and chemical hydrolysis of bensulfuron-methyl. Weed Research 36, 441-448.
Sarmah AK (1998) Persistence and mobility of triasulfuron, metsulfuron-methyl and chlorsulfuron in alkaline soils. PhD thesis, The University of Adelaide.
Sarmah AK, Kookana RS (1999) Simultaneous analysis of traisulfuron, metsulfuron-methyl and chlorsulfuron in water and alkaline soils by high performance liquid chromatography. Journal of Environmental Science & Health- Part B: Pesticides, Food Contaminants, & Agricultural Wastes 34, 363-380.
Sarmah AK, Kookana RS, Alston AM (1998) Fate and behaviour of triasulfuron, metsulfuron-methyl, and chlorsulfuron in the Australian soil environment: a review. Australian Journal of Agricultural Research 49, 775-790.
Sarmah AK, Kookana RS, Alston AM (1999) Degradation of chlorsulfuron and triasulfuron in alkaline soils under laboratory conditions. Weed Research 39, 83-94.
Starr JL, Glotfelty DE (1990) Atrazine and bromide movement through a silt loam soil. Journal of Environmental Quality 19, 552-558.
Steenhuis TS, Staubitz W, Andreini MS, Surface J, Richard TL, Paulsen R, Pickering NB, Hagerman JR, Geohring LD (1990) Preferential movement of pesticides and tracers in agricultural soils. Journal of Irrigation and Drainage Engineering 116, 50-56.
Stork P (1995) Field leaching and degradation of soil applied herbicides in a gradationally textured alkaline soil: chlorsulfuron and triasulfuron. Australian Journal of Agricultural Research 46, 1445-1458.
Strek HJ (1998) Fate of chlorsulfuron in the environment. 2. Field evaluations. Pesticide Science 53, 52-70.
Tillman RW, Scotter DR, Clothier BE, White RE (1991) Solute movement during intermittent water flow in a field soil and some implications for irrigation and fertilizer application. Agricultural Water Management 20, 119-133.
Trojano J, Garretson C, Krauter J, Huston J (1993) Influence of amount and method of irrigation water application on leaching of atrazine, Journal of Environmental Quality 22, 290-298.
Vicari A, Catizone P, Zimdahl RL (1991) Bioactivity, degradation and mobility of chlorsulfuron in soil. Rivista di Agronomia 3, 400-406.
Vicari A, Catizone P, Zimdahl RL (1994) Persistence and mobility of chlorsulfuron and metsulfuron under different soil and climatic conditions. Weed Research 34, 147-155.
Vischetti C, Businelli M (1992) Evaluation of a simulation model for prediction of chlorsulfuron fate in an Umbrian soil. Science of Total Environment 123/124, 561-569.
Walker A, Cotterill EG, Welch SJ (1989) Adsorption and degradation of chlorsulfuron and metsulfuronmethyl and triasulfuron in soils from different depths. Weed Science 29, 281-287.
Walker A, Welch SL (1989) The relative movement and persistence in soil of chlorsulfuron, metsulfuron-methyl and triasulfuron. Weed Research 29, 375-383.
Walker S, Robinson GR (1996) Chlorsulfuron residues are not accumulating in soils of southern Queensland. Australian Journal of Experimental Agriculture 36, 223-228.
White RE (1985) The influence of macropores on the transport of dissolved and suspended matter through soil. Advances in Soil Science 3, 95-120.
Manuscript received 25 June 1999, accepted 15 November 1999
A. K. Sarmah(ABD), R. S. Kookana(AC), and A. M. Alston(AB)
(A) Cooperative Research Centre for Soil and Land Management, PMB 2, Glen Osmond, SA 5064, Australia.
(B) Department of Soil and Water, The University of Adelaide, PMB 1, Glen Osmond, SA 5064, Australia.
(C) CSIRO Land and Water, PMB 2, Glen Osmond, SA 5064, Australia.
(D) Corresponding author and present address: School of Civil Engineering, 1284 Civil Engineering Building, Purdue University, West Layfayette, IN 47902-1284, USA.