JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 27 June 2007
Published in J Environ Qual 36:1205-1216 (2007)
DOI: 10.2134/jeq2006.0241
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
Related Collections
Right arrow Preferential Flow
Right arrow Pesticides
Right arrow Organic Compounds
Right arrow Other Environmental Contamination

TECHNICAL REPORTS

Vadose Zone Processes and Chemical Transport

Micro-Trench Experiments on Interflow and Lateral Pesticide Transport in a Sloped Soil in Northern Thailand

G. Kahl*,a, J. Ingwersena, P. Nutniyoma, S. Totrakoolb, K. Pansombatc, P. Thavornyutikarnd and T. Strecka

a Univ. of Hohenheim, Inst. of Soil Science and Land Evaluation, Biogeophysics Section, D-70599 Stuttgart, Germany
b EHWM CU/CMU, Chiang Mai Univ., 50200 Chiang Mai, Thailand
c Chiang Mai Univ., Faculty of Agriculture, 50200 Chiang Mai, Thailand
d Chiang Mai Univ., Faculty of Science, 50200 Chiang Mai, Thailand

* Corresponding author (gunnarkahl{at}hotmail.com)

Received for publication June 26, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
During recent decades, a change in land use in the mountainous regions of Northern Thailand has been accompanied by an increased input of agrochemicals. We identified lateral water flow and pesticide transport pathways and mechanisms in a Hapludult on a sloped litchi orchard in Northern Thailand. During two rainy seasons, two micro-trench experiments were performed at the plot scale (2 by 3 m). The first experiment was performed at the footslope of the orchard; the second was performed at a midslope position. Two salt tracers (bromide and chloride) and two pesticides {methomyl [S-methyl-N-(methylcarbamoyloxy)thioacetimidate] and chlorothalonil (2,4,5,6-Tetrachlor-1,3-benzdicarbonitril)} were applied in stripes parallel to the slope 150 and 300 cm away from the trench. At the trench, soil water was collected by wick samplers. Tensiometers and time-domain reflectometry probes were installed. At the end of the experiment, soil samples were taken and analyzed for residual concentrations of tracers and pesticides. Lateral subsurface flow of water occurred exclusively along preferential flow paths and was mainly observed at 0- to 30- and 60- to 90-cm depth. Lateral transport of pesticides was negligible, but both pesticides were found beneath the application area at 90 cm depth. Therefore, they may pose a groundwater contamination risk. The amount of wick flow and the location of interflow were mainly a function of rain amount and antecedent soil water suction. During dry periods, water flow was restricted to the topsoil. After heavy rain events and wet periods, interflow was mainly observed in the subsoil. The cumulative rain amount between samplings necessary to induce interflow was 20 mm. At the footslope, the interflow was seven times higher, and the network of water-bearing pores increased compared with the midslope position.

Abbreviations: ACN, acetonitrile • AEC, anion exchange capacity • HPLC, high-performance liquid chromatography • MeOH, methanol


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
LARGE PARTS of the mountainous regions of Southeast Asia are undergoing tremendous changes in land use. In Northern Thailand, triggered by high population growth and migration from the lowlands, agricultural areas are expanding into ecologically more vulnerable regions. The area used for litchi production, for example, was extended from 17 224 to 22 937 ha between the years 1996 and 2000, with more than 60% of the area in the two northern provinces of Chiang Mai and Chiang Rai. About 90% of the fruits are produced in the uplands of Northern Thailand (Huang et al., 2005). Litchi was introduced to Thailand more than 100 yr ago (Papademetriou and Dent, 2002). It ranks eleventh in value of fruit production in Thailand. The total litchi output increased from 55 355 to 81 388 Mt during the same period (Menzel, 2002). The export earnings from litchi increased between 1991 and 2002 at a rate of 10% per year. High-quality criteria for the export and the rapid development of resistance of insects led partly to an insecticide overuse by farmers (Rerkasem, 2004). In some regions, local farmers have been found to be highly poisoned by pesticides (Petchuay et al., 2006; Stuetz et al., 2001). Agrochemicals leach to groundwater and river waters and probably contribute to the contamination of surface water and drinking water (Chusaksri et al., 2006; Kruawal et al., 2005; Kumblad et al., 2001) in the lowlands. For Northern Thailand, only a study by Ciglasch et al. (2005) has addressed the problems of pesticide leaching in an upland soil. At the same study site used in the present work, they observed that, after a heavy rain (80 mm), about 1% of the applied mass of several pesticides was leached below 0.55 m. The fast vertical transport of different chemicals through upper soil layers has been recorded in various soils (Bundt et al., 2000; Kladivko et al., 2001; Stamm et al., 2002). On a clayey till in Denmark, for example, chloride and several organic compounds were found at a depth of 250 cm within less than 1 d after application (Broholm et al., 2000; Nilsson et al., 2001; Sidle et al., 1998). Flow took place along a complex interconnected system of fractures and macropores. Flury (1996) reviewed the literature on pesticide transport through soil at the field scale. He estimated that about 0.1 to 5% of the applied mass of pesticide is displaced below the root zone, depending on the timing of heavy rain after application.

Although the vertical preferential transport has been investigated intensively over the last two decades, lateral preferential transport has received much less attention. Lateral flow or solute transport on sloped land stands for movement downhill. Particularly in strongly sloped areas, lateral subsurface transport of pesticides may be an important process. Preferential lateral water flow and solute transport can take a variety of forms, including movement along the soil–rock interface (Buttle and McDonald, 2002; Collins et al., 2000; Peters et al., 1995), runoff along microchannels above the bedrock surface (Noguchi et al., 1999), pipe flow at the base of the soil profile (McDonnell, 1990), flow through a self-organizing interconnection of macropores and mesopores embedded in the soil matrix (Sidle et al., 2001; Sidle et al., 2000), or as lateral (near-) surface runoff (Stamm et al., 2002). The main factors influencing lateral interflow are rain amount, antecedent soil moisture, and antecedent precipitation (Sidle et al., 1995; Sidle et al., 2000; Tromp-van Meerveld and McDonnell, 2006b; Uchida et al., 2002).

To explore lateral flow pathways, several methodologic approaches have been used in the past. These approaches include the direct measurement of water flux (Tromp-van Meerveld and McDonnell, 2006a) and the monitoring of the transport of ideal tracers like bromide and chloride (Collins et al., 2000; Feyen et al., 1999) or dye tracers (Noguchi et al., 1999; Stamm et al., 2002).

We know little about the mechanisms of lateral pesticide transport. Johnson et al. (1996) studied the relative importance of surface-, lateral subsurface-, and drainage transport of isoproturon [3-(4-isopropylphenyl)-1,1-dimethyluree] in a moderately sloped clay soil in Oxford, UK, during two major rain events. They stated that 23% of a total recovery of 0.7% of mass was transported with lateral subsurface flow. Brown et al. (1995) studied the behavior of isoproturon over a 2-yr period on a similarly sloped and drained clay soil in Northumberland, UK. They found a maximum 0.45% of applied isoproturon in the lateral flow within the first 30 cm soil depth, but the experimental setup did not distinguish between surface and subsurface transport. Truman et al. (1998) reported a recovery of 6.2% of the applied amount of Fenamiphos [Ethyl-3-methyl-4-(methylthio)phenyl(1-methylethyl)phosphoramidat] in the lateral interflow collected at 100 cm depth in a moderately sloped loamy sand in Georgia. During the 3-yr period, only 0.1% of Fenamiphos occurred in the surface runoff. To our knowledge, no detailed description of lateral subsurface transport of pesticides in tropical soils is available.

The present study explored the mechanisms and pathways of lateral water flow and solute transport in a sloped kaolinitic soil in Northern Thailand. In micro-trench experiments, we applied two salt tracers, the insecticide methomyl and the fungicide chlorothalonil, and monitored concentrations in interflow and soil.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Research Area
The study area is located 30 km northwest of Chiang Mai, Northern Thailand, at an altitude of 800 m a.s.l. The site is used as a litchi (Litchi chinensis Sonn.) orchard. Downslope, the orchard borders on a creek (Fig. 1). The soil type is kaolinitic, thermic Hapludult (Ciglasch et al., 2005) over the parent material granite. Soil depth ranged from more than 150 cm at the plateau position to over 100 cm at a slope of 15° and 100 cm at a slope of 25° and 40 cm at 35°. The research plot at the footslope had an inclination of 22°, and the midslope plot had an inclination of 23°. A detailed soil description, including the areal density of macropores (diameter >1 mm), is given in Table 1. Macropores with diameters between 1 and 10 cm were observed mainly at a depth of 60 to 90 cm during installation of the experiment. All macropores at the investigated soil profiles had diameters less than 0.5 cm. The lower boundary of termite nests, which form large burrows, was at around 95 cm depth. A weather station was installed at the research site and provided data for temperature and precipitation (except rain from 6–29 Sept. 2005; daily values are available from another weather station in the same watershed) at a resolution of 10 min. Average annual precipitation and temperature are 1500 mm and 22°C, respectively. The rainy season starts in late May and ends in September, followed by a cool dry season from October to February and a hot dry season from March to April (Fig. 2).


Figure 1
View larger version (101K):
[in this window]
[in a new window]

  		Fig. 1. Sketch of the study site. Filled circles: Litchi trees.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Soil properties.

 

Figure 2
View larger version (52K):
[in this window]
[in a new window]

  		Fig. 2. Monthly precipitation and air temperature at the study site.

 
Micro-Trench Experiments
Two micro-trench experiments (2 by 3 m) were performed, one in 2004 on a plot at the footslope close to the river and one in 2005 on a plot about 50 m above the former plot. In this article, the latter position is termed "midslope" (Fig. 1). Three sides of the plot were excavated to a depth of 1.5 m if possible or to the parent material but at least to 1.2 m (Fig. 3). The upslope side remained undisturbed to allow water influx from above. At four distances from the trench (75, 125, 175, and 225 cm) and at six depths (15, 35, 50, 75, 100, and 120 cm), tensiometers and time domain reflectometry probes were installed in the side walls. At the downhill profile, in each depth of 10 to 20, 40 to 50, 70 to 80, and 100 to 110 cm, we installed six capillary fiber glass wick samplers (Pepperell Braiding Company, Pepperell, MA). The ends of the wicks were unfurled to a circle with a diameter of about 8 cm on stainless steel sheets, which were pressed firmly to the soil profile. Wick samplers were therefore connected to macropores and soil matrix. The plaited end of the wicks hung out of a hole in the metal plates. The wick had a filament diameter of 0.7 µm, a filament tortuosity of 0.79, a porosity of 0.90, a bulk density of 0.25 g cm–3, a diameter of 1.45 cm (Knutson and Selker, 1994), and a hydraulic conductivity of 10–2 cm s–1 (Holder et al., 1991). Numerous studies have shown that capillary wick samplers are suitable to collect soil water and solutes (BrandiDohrn et al., 1996; Goyne et al., 2000; Louie et al., 2000). Holder et al. (1991) found that adsorption of inorganic ions and selected organic chemicals to this wick material was very small. In the present study, preliminary experiments showed that sorption of the studied chemicals to the wicks was negligible (data not shown). Boll et al. (1992) investigated matric potentials inside 50-cm-long wicks at different flow rates. The suctions ranged between 3 and 50 cm. Wick outflow was collected continuously, and the sampling containers were changed every 1 to 2 d. The contact area between the wick and the soil was about 40 cm2. The area covered by the wicks was 4% of the soil profile. We present some integrated values for anion and pesticide findings in the discussion section. Data of the six wicks (240 cm2) of each depth were integrated over an area of 200 by 30 cm.


Figure 3
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 3. Setup of the plot experiments. Application area I: at the footslope Br–, at midslope Cl–; application area II: at the footslope Cl–, at midslope Br–, methomyl, and chlorothalonil.

 
The two pesticides—methomyl (commercial grade: E.I. Du Pont [Wilmington, DE]; standard: Riedel de Haen, Germany) and chlorothalonil (commercial grade: Cafforo, S.p.A., Italy; standard: Riedel de Haen, Germany)—were chosen for the following reasons: (i) They are very different with regard to their sorption properties; (ii) they had not been applied at the study site in recent years, which was especially important to avoid background noise; and (iii) they could be analyzed by high-performance liquid chromatography (HPLC). Both pesticides and the salts KBr and NaCl (Merck Ltd., Darmstadt, Germany) were applied with 2 L of water to two slope-parallel stripes of 10 cm width. Dates, amounts, and locations of tracer applications are given in Table 2. The measurement period in 2004 started on 5 June and ended on 14 September, and in 2005 it started on 8 June and ended on 5 October. At the midslope in 2005, methomyl and the salts were applied on the same plot three times and chlorothalonil twice. In the following discussion, the depth of 0 to 30 cm is referred to as depth 1, 30 to 60 cm as depth 2, 60 to 90 cm as depth 3, and 90 to 120 cm as depth 4. Soil samples were taken at different positions (footslope: 20, 60, 100, 130, 140, 150, 175, 195, 205, 230, 270, and 280 cm distance from the trench; midslope: between 100 and 160 cm and 260 and 310 cm, in 10-cm resolution) using a Pürckhauer auger to 1 m depth. Each sample was divided into 10 subsamples, each representing 10 cm soil depth. The samples were analyzed for their residual tracer and pesticide concentrations. Recovery of the residual solute concentrations was estimated by integrating the concentrations of the solute plumes over the covered area (McCord et al., 1991). For visualization, the tracer data were interpolated over the longitudinal cross-section of the plot using the inverse distance weighting method with a power value of 2.0.


View this table:
[in this window]
[in a new window]

  		Table 2. Application modality.

 
Analytical Procedures
Methomyl and Chlorothalonil
To extract methomyl from soil, 5 g of dried soil was horizontally shaken with 12.5 mL acetonitrile (ACN) (Merck) for 3 h and centrifuged (4500 g for 10 min). The supernatant was kept in a 25-mL volumetric flask. The procedure was repeated once. The volume of the composite sample was adjusted to 25 mL with ACN. The solution was filtered by nylon syringe filters (0.45 µm) (Chrom Tech, Apple Valley, MN), 5 mL pipetted into a glass vial, dried in a gentle stream of nitrogen, and rediluted with 3 mL ACN/methanol (MeOH) (Merck)/H2O (ratio 35/30/35 solution by volume). Pesticide recovery using this method was 86%.

Extraction of chlorothalonil from soil was optimized based on USEPA Method 3550C (USEPA, 2000). Dried soil (3 g) was sonicated for 20 min with 10 mL dichloromethane (Labscan Asia Co., Bangkok, Thailand)/ACN (50/50 solution by volume) and manually shaken every 3 min for 15 s. After centrifuging (4500 g for 5 min), the supernatant was placed in a 25-mL volumetric flask. This procedure was repeated once. The volume of the composite sample was filled up to 25 mL with dichloromethane/ACN (50/50 solution by volume). A volume of 10 mL was filtered with a 0.7-µm glass fiber filter (GF/F) (Whatman Inc., Clifton, NJ). Finally, 3 mL of the filtered solution was pipetted into a 10-mL vial. The solvent was dried with nitrogen, and the remains were rediluted with 3 mL ACN/H2O (80/20 solution by volume). Recovery using this procedure was 95 to 103%.

Water samples to be analyzed for methomyl were poured through a nylon filter (PVDF W/GMF) (Whatman) (Masque et al., 1998). Samples were extracted by solid-phase extraction (FLORISIL, tube size 3 mL; Restek, Bellefonte, PA). To activate the sorbent, the cartridge was rinsed with 3 mL ACN/MeOH (50/50 solution by volume) followed by 3 mL MeOH and 6 mL of distilled water. In the extraction step, 10 mL of the sample was applied to the sorbent at a drop-by-drop rate. A volume of 150 mL was extracted. For some samples, the collected volume was less than 150 mL. Pretests showed that the efficiency of the cartridges did not vary with sample volume. In the next step, the cartridge was dried for 2 min under suction to remove the remaining water. Finally, 3 mL of ACN/MeOH/H2O (35/30/35 solution by volume) were used to elute methomyl from the sorbent. Samples analyzed for chlorothalonil were poured through a glass fiber filter. The extraction step was performed with C18–cartridges (tube size 3 mL) (Restek) activated with 5 mL MeOH and 5 mL of distilled water. The elution step was done with 3 mL ACN. Extracted solutions were kept in dark glass bottles at –20°C. The recovery using solid-phase extraction was 94% for methomyl and 96% for chlorothalonil.

Pesticide samples were analyzed using HPLC equipped with a photo-diode-array detector (Shimadzu, Japan) and an ultracarbamate column (C18, 5 µm, 4.6 by 250 mm) (Restek). The mobile phase was for the HPLC analysis of methomyl ACN/MeOH/H2O (35/30/35 solution by volume) and for chlorothalonil ACN/H2O (80/20 solution by volume). The flow rates used for methomyl and chlorothalonil analyses were 1.0 mL min–1 and 1.2 mL min–1, respectively. The wavelengths used for methomyl and chlorothalonil were 233 nm and 231 nm, respectively. The injection volume was 50 µL. Limits of detection of methomyl and chlorothalonil were 2 µg L–1 and 3 µg L–1, respectively.

Degradation Experiment
To determine degradation of methomyl, 5 g of air-dried soil (<2 mm) was weighed into a 50-mL flask, and 0.5 mL of stock standard solution was added. The organic solvent was evaporated at room temperature. Distilled water was added to reach the desired soil moisture. The flask was incubated at the desired temperature, and samples were taken at Day 0, 5, 10, 15, 20, 30, 40, and 60 after pesticide application. The flask was weighed regularly, and distilled water was added to maintain soil moisture. Methomyl was analyzed as described previously.

Bromide and Chloride
To measure bromide and chloride in the soil, samples were shaken with water (20 g/20 mL) for 3 h and centrifuged (4500 g for 5 min). The supernatant was decanted and passed through a 0.45-µm cellulose-nitrate filter (Whatman). Further details are given in Kumke et al. (1999). Tracer concentrations were analyzed by ion chromatography (DX 300; Dionex Inc., Sunnyvale, CA). The ion chromatograph was equipped with a AS9-HC column (4 by 250 mm) (Dionex) and a AG9-HC precolumn (4 by 50 mm) (Dionex). The retention times of Cl– and Br– were 6.92 and 10.77 min, respectively. Both anions could be well separated by this method.

Measurement of Sorption Isotherms
Soil samples for pesticide sorption and degradation experiments were taken across the research field on a regular grid at depths of 0 to 0.15, 0.15 to 0.3, and 0.3 to 0.6 m. The soil was air dried and sieved to less than 2 mm. For the pesticide sorption experiment, 100 mg L–1 methomyl or chlorothalonil stock solution in methanol and acetonitrile, respectively, was used to prepare five solutions with concentrations of 0.15, 0.45, 0.75, 1.05, and 1.35 mg L–1. The organic solvent was dried in a gentle N2 stream before methomyl was redissolved with 0.01 M CaCl2 solution. We added 20 mL of standard solution to 10 g of soil in a 50-mL Teflon tube. Samples were shaken for 24 h at room temperature and centrifuged for 15 min at 4500 rpm. The supernatant (10 mL) was removed and stored in a glass tube at 4°C for further extraction and HPLC analysis.

The Freundlich equation was fitted to the sorption data:

Formula 1[1]
where C (mg L–1) is the concentration of dissolved pesticide, S (mg kg–1) is the sorbed phase concentration, k is the Freundlich coefficient (mg1–m Lm kg–1), and m is the Freundlich exponent.

To establish Cl– and Br– sorption isotherms and to determine the anion exchange capacity (AEC) of the soil, samples were taken from a soil pit close to the midslope plot at depths of 0 to 0.15, 0.15 to 0.3, 0.3 to 0.6, 0.6 to 0.9, and 0.9 to 1.2 m. The samples were air dried and sieved to less than 2 mm. Sorption isotherms were established based on Collins et al. (2000). A volume of 20 mL of NaCl– or KBr solutions in concentrations of 0.017, 0.17, 1.7, 8.5, and 17.1 mmol L–1 was shaken for 24 h with 10 g soil and centrifuged (3000 g for 30 min). The supernatant was analyzed for the Cl– and Br– content. The isotherms were corrected for the residual salt concentrations, which were determined by shaking 5 g of soil with 20 mL 0.005 M MgSO4 for 12 h, centrifuging the samples, and analyzing the supernatant.

Anion Exchange Capacity
The AEC (cmolc kg–1) of the soil was measured using the method of Gillman and Sumpter (1986). We placed 2 g of soil into a centrifuge tube. The tube was weighed (W1 [g]) and shaken with 20 mL 0.1 M BaCl2/0.1 M NH4Cl solution for 2 h and centrifuged for 30 min (3000 rpm), and the supernatant was discarded. The soil was washed once with 20 mL of 0.05 M BaCl2 and three times with 0.002 BaCl2. The pH of the solution at the last washing step was measured and adjusted to original soil pH with 0.1 M HCl. The samples were allowed to stand overnight. Samples were centrifuged, and the supernatant was analyzed for the Cl– concentration (C1 [mmol mL–1]). The tubes were weighed (W2 [g]), and 10 mL 0.005 M MgSO4 solution was added. The samples were shaken for 2 h, and the pH value was adjusted with 0.1 M H2SO4 to the original pH. Samples were weighed (W3 [g]) and allowed to stand overnight. After the samples were centrifuged, the supernatant was analyzed for the Cl– concentration (C2 [mmol mL–1]). AEC was calculated as

Formula 2[2]

Formula 3[3]

Formula 4[4]
where M is the soil in weight (g).

Estimation of Transport Depths
The retardation factor R for chloride and bromide was calculated as

Formula 5[5]
where {rho}s is the bulk density of the soil (g cm–3), Kd is the linear sorption coefficient, and {theta} is the volumetric water content. The retardation factor R was calculated with averaged measured water contents per depth and a soil density of 1.2 g cm–3 (Spohrer et al., 2006a). Transport depths of the tracers were estimated as

Formula 6[6]
where Z is the transport depth (cm), and Q is the cumulative rain (cm) over the measurement period. For simplicity and because the experiments took place in the rainy season, we used the potential evapotranspiration (ETpot), which was calculated as

Formula 7[7]
where ET0 is the potential reference evapotranspiration based on calculations from Clarke et al. (1992), and Kc is the evapotranspiration coefficient. For our research site, Spohrer et al. (2006b) found Kc values ranging between 1 and 0.45 depending on the wetness of the soil surface. For the rainy season, Spohrer et al. (2006b) give an average Kc value of 0.8, which results in a daily ETpot of 3.1 mm d–1.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSION
 REFERENCES
 
Rainfall
In 2004, the total precipitation during the measurement period was 794 mm; in 2005, it was 888 mm. For the following analysis, rain data were divided in rain events. If the time between two rainfalls was larger than 1 h, they were treated as two different rain events. The major rain event in 2004 occurred on 15 June (62.2 mm; maximum 10-min intensity: 91.8 mm h–1) and was sampled together with a 17.7-mm event on 16 June. In 2005, the biggest event occurred on 13 August (61.3 mm; maximum 10-min intensity: 19.8 mm h–1), but it produced no wick outflow.

Soil Water Flux
Soil water content by volume averaged over the measurement period and all time domain reflectometry probes was 35% at the footslope (year 2004) and 21% at midslope position (year 2005). In both years, only small spatial fluctuations of the water content and suction were observed (data not shown). Therefore, for further evaluation, only the average of soil water suction taken over all depths and locations was used. At midslope position, the water suction was, on average, 33 hPa (= centimeter water column) (SD 23 hPa) and at the footslope was 30 hPa (SD 37 hPa).

Wick Outflow
At both slope positions, wick outflow was mainly observed in the first and third depth. At the footslope, wick outflow greatly differed within these two depths. In contrast, wicks in the second and fourth depth collected similar volumes of water (Fig. 4A). At midslope position, only 8 of 24 wicks collected water (Fig. 4B); six of them were located in the first and third depth. With 7.7 L, the wick outflow at midslope position was about seven times lower than at the footslope. Summed up, the first and third depth leached 62% of the total outflow at the footslope and 86% at the midslope position. The SD of water volume collected per wick was highest in the first (0.3 L) and in the third (0.2 L) depth at the footslope and in the third (1.7 L) and fourth (0.4 L) depth at the midslope position.


Figure 4
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 4. Cumulative outflow per wick (A) at the footslope from 5 June to 14 Sep. 2004 and (B) at midslope position from 8 June to 5 Oct. 2005.

 
Generally, wick outflow was collected directly after heavy rain events, indicating that the subsurface water flow reacts very fast. To trigger wick outflow, the rain amount had to exceed 20 mm (data not shown). In both years, we observed that after a dry period a strong rain event initially induced only little wick outflow, whereas subsequent smaller rain events yielded a much higher outflow (Fig. 5A and 5B). The storm event (61 mm) on 13 Aug. 2005, which ended a dry period, did not produce wick flow. A similar rain event (62 mm) on 15 June 2004, which infiltrated into a wet soil, produced a wick outflow of 14.3 L, which corresponds to 25% of the outflow of the whole measurement period.


Figure 5
View larger version (44K):
[in this window]
[in a new window]

  		Fig. 5. Total volume of wick outflow (A) at the footslope and (B) at midslope position. The letters a, b, and c indicate wick outflow events induced by rain events of decreasing magnitude.

 
If wick outflow occurred under dry conditions, it was generated mainly by the wicks of the first depth. At midslope position, the main wick outflow in the first depth was observed when the average water suction rose above 40 cm on the sampling date, whereas flow in the third depth mainly occurred during periods with water suctions below 40 cm (Fig. 6).


Figure 6
View larger version (36K):
[in this window]
[in a new window]

  		Fig. 6. Relative wick outflow of depth 1 and 3 at the midslope position. The soil water suction was shifted 1 d ahead, therefore representing the antecedent conditions.

 
For sampling dates when wick outflow occurred, the effect of total rain amount between the sampling dates (for both measurement periods together) and water suction on the wick outflow volume could be described by the regression equation

Formula 8[8]
where V is the wick outflow volume (L), h represents soil water suction (cm) measured at the sampling date, R denotes the rainfall amount between the sampling dates (cm), and a and b are dimensionless empirical coefficients. The coefficients a and b were estimated after log transformation of Eq. 8 by nonlinear regression as 0.32 and 1.18 (R2 = 0.54) based on the data of all sampling dates at which outflow occurred at both measurement periods. Substituting the soil water suction by the 1-h rain intensity reduced the R2 to 0.12.

Solute Transport
Residual Concentrations
Under lab conditions (30°C and a water content of 32% by weight), the average half-life of methomyl in the topsoil was 15.5 d, which is in the range of published values (Extoxnet, 1996). Under field conditions, however, at the same research site Ciglasch et al. (2006) found significantly decreased field half-lives of several pesticides compared with published values. Therefore, we might expect a faster decay of methomyl and chlorothalonil. Published values for chlorothalonil range from 1 to 3 mo (Extoxnet, 1996). The measured pesticide sorption proved that methomyl sorption (k value, 0.12–9.09 mg1–m Lm kg–1; m, 0.96–1.61 in different depths) was much weaker than that of chlorothalonil (k value, 118.8–401.5 mg1–m Lm kg–1; m, 0.35–0.49).

No chloride sorption was observed in the topsoil (0–30 cm). In subsoil (30–120 cm), chloride sorption increased with depth. Sorption of bromide was stronger than that of chloride, and, in contrast to chloride, bromide was weakly sorbed in the topsoil. Similar to chloride, sorption of bromide increased with soil depth. This finding is in accordance with the measured AEC, which was lowest in the topsoil (0.1 cmolc kg–1) and highest in the subsoil (0.4 cmolc kg–1). The estimated transport depths taking into account retardation were estimated to be 92 cm for chloride (measured transport depth, 75 cm) (Fig. 7) and 54 cm for bromide (measured transport depth, 85 cm) in 2004. In 2005, after three subsequent tracer applications, the chloride peaks were estimated to be located between 60 and 120 cm depth (measured transport depth, 90 cm) and bromide between 30 and 65 cm (measured transport depth, 60 cm).


Figure 7
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 7. Residual concentrations of chloride and bromide in the soil solution at the footslope, (A and B) and at midslope position (C and D), and extracted residual methomyl and chlorothalonil concentration at midslope position, including sorbed pesticide masses (E and F). The arrows mark the position of application; the line marks the vertical transport direction.

 
The residual concentrations in soil at the end of the experiment were highest within 20 cm distance downhill (Fig. 7). At midslope position, the amounts of recovered bromide were higher than at the footslope (Table 3). For chloride, the recovery was similar at both slope positions. The pesticide residues were much smaller (methomyl 0.1%; chlorothalonil 7%) than the found mass of tracers. Although the main chlorothalonil peak remained in the first 20 cm of the soil, the highest methomyl concentrations were found in 90 cm, but in very low concentrations (Fig. 7E and 7F).


View this table:
[in this window]
[in a new window]

  		Table 3. Residual tracer and pesticide recovery.

 
Wick Outflow
In total, at the footslope about 10 mmol and at the midslope about 0.65 mmol chloride were detected in the wick outflow. Detected bromide was about 0.66 mmol at the footslope and 0.02 mmol at the midslope position (Table 4). In both years, chloride and bromide reached the profile wall at the same time, although the application areas were located in different distances to the profile wall (1.5 and 3 m) (Fig. 8). At the footslope, more than 90% of the detected salts were transported with the first heavy rain event in all depths (first flush). Later, considerable concentrations of the tracers were found only in the third depth. At midslope position, both tracers occurred first in the wick outflow of the first depth and later in that of the third and fourth depth. In contrast to the footslope, they were found throughout the measurement period.


View this table:
[in this window]
[in a new window]

  		Table 4. Chloride, bromide, and pesticide recovery in interflow. Applied masses were 2.5 mol in the year 2004 and 5 mol in 2005.

 

Figure 8
View larger version (49K):
[in this window]
[in a new window]

  		Fig. 8. Cl– and Br– recovery at the soil profile at the footslope (A and B) and at midslope position (C and D) and rain amount between the samplings. The arrows mark the dates of the second and third application.

 
Pesticide amounts in the wick outflow were low at both slope positions (Table 4). Although chlorothalonil was detected seven times, exclusively in samples of the first and third depth, methomyl concentrations were detected 16 times (Fig. 9), including twice in the fourth depth. During the first application period from 8 June 2005 to 3 Aug. 2005, methomyl was found in the same samples as chloride. In the beginning, methomyl was detected in the outflow of the wicks in the first depth and later in the outflow of the wicks of the third depth. After the second application in August 2005, the main transport was observed in the third and in the fourth depth. Chlorothalonil was not detected in the outflow during the first month after application. Leaching of chlorothalonil started during a period of strong rainfalls in September. It was initially found in the topsoil and subsequently in the third soil depth. In contrast to methomyl, the chlorothalonil transport in the first soil depth was higher than in the third. The total mass of chlorothalonil found in the wick outflow was only about 3% of that of methomyl.


Figure 9
View larger version (38K):
[in this window]
[in a new window]

  		Fig. 9. (A) Methomyl and (B) chlorothalonil recovery at midslope position. The arrows mark the dates of the second and third application.

 
Solute transport over a distance of 1.5 m mainly occurred in the first depth and over a distance of 3 m in the third depth (Table 4).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSION
 REFERENCES
 
Are Lateral Water Flow and Solute Transport Preferential?
Soil water was collected by wick samplers. Dry wicks may de-water pores of small diameters due to higher suction. The intruding water gradually decreases the suction of the wick material to the point of saturation. At this point, the wick starts to release water to the sampling bottle. The suction at saturation is controlled by the hanging water column, which was 15 cm. This suction is able to empty pores of effective diameters greater than 1 mm. Pores of this diameters are often classified as macropores (Luxmoore et al., 1990), although Beven and German (1982) stated that macropores are structures that allow channeling flow, whatever their size. When the wicks are saturated, they contain about 1 mL of water per centimeter of wick. Therefore, in an unsaturated soil, pores with diameters less than 1 mm contribute at maximum 15 mL water per sample. However, the effect of some water backing up at the profile wall before saturation has to be taken into account. Our data did not provide evidence for such a "backing-up" effect. Water suction measured at the nearest position to the soil profile (75 cm distance) was not decreased (data not shown). We therefore conclude that water collected by the wick samplers is mainly generated by macropores.

Several studies reported under- or overestimation of water fluxes of up to 34% relative to the water balance (BrandiDohrn et al., 1996; Louie et al., 2000), depending on the differences in matric potential between wicks and soil. The integrated values of solute recoveries given later should be considered to be rough estimates. With the exception of chlorothalonil, all simultaneously applied chemicals reached the soil profile at the same time, regardless of their different application distances from the soil profile. This can only be explained by long preferential flow paths that transport solutes over several meters during one rain event.

The presence of the residual concentrations of tracers and pesticides directly beneath the application area indicates that matric water flow and solute transport was mainly vertical. The lateral (or downhill) transport distance in the soil matrix was only 0 to 20 cm. We conclude, therefore, that the chemicals found in the wick outflow were transported by preferential flow.

The highest residual methomyl concentrations were found at a soil depth of 80 cm. The main masses (in the outflow and soil matrix) of chlorothalonil were found in the upper soil. However, even this strongly sorbing chemical was found in the outflow of the third depth shortly after application. Such long transport distances can only be explained by preferential transport. Both pesticides probably pose a threat to river water and groundwater quality.

Recovery of chloride and bromide in the total wick outflow ranged between less than 1 and 4%. These estimates are uncertain. They are in the range reported for a soil profile of a groundwater-influenced clay in Switzerland by Feyen et al. (1999). These authors usually recovered in the interflow water (5–30 cm depth) between less than 1 and 5% of chloride and bromide applied at the soil surface. Only once was the recovery significantly higher (22%; experiment in which no fast breakthrough occurred).

Which Factors Influence Lateral Water Flow and Solute Transport?
Our finding that wick outflow at the footslope was seven times higher than at the midslope supports the hypothesis that the processes governing preferential interflow strongly depend on slope position. Water from the whole uphill microcatchment might have contributed to wick outflow. Zehe and Flühler (2001) found that the soil at the bottom of a slope had a higher susceptibility for preferential flow than soil anywhere else.

Further observations support the importance of the soil moisture. At both slope positions, there were periods when, under dry conditions, a strong rain event created minimal wick outflow, although subsequent weaker events led to higher wick outflow. This shows that, at similar rainfall amounts, a dry soil delivers less interflow than a wet soil. This can also be seen when comparing the two major rain events during both measurement periods (both around 60 mm): Wick outflow was high at the footslope during wet conditions, whereas no wick reacted at midslope during dry conditions. Thus, at midslope, the buffer capacity of the soil was high, and the rainwater was stored in the soil matrix. At the footslope, the buffer was low, and the percolating water led to great wick outflow. The regression analysis of lateral subsurface flow demonstrates that wick outflow is mainly driven by soil water suction and by rainfall amount. Abundant rain and low water suction lead to a high wick outflow. Numerous studies stress that rain intensity greatly affects lateral subsurface flow (Haga et al., 2005; Uchida et al., 2005). This could not be confirmed by our work.

Interflow was collected mainly at 0 to 30 cm and 60 to 90 cm at both slope positions. This flow pattern seems to be typical for the study site. In an earlier study 50 m further uphill, Spohrer et al. (2003) had found interflow at similar depths. Soil water suction controlled in which of those two depths wick outflow occurred. When the average suction was greater than 40 cm, as was the case several times at the midslope position, interflow occurred mainly in the first depth. After rewetting, wick discharge occurred mainly in the third depth. At the footslope, this behavior was not observed because a suction drop occurred only in the upper soil depth, whereas the lower soil remained wet (data not shown). A similar observation was made by Gaskin et al. (1989), who found water flow in the A-horizon under dry conditions and in the B-horizon under wet conditions.

The effect of soil moisture on the extent of preferential interflow has been observed at various study sites. In an intensively researched watershed in Japan, lateral preferential flow paths were shorter than 62 cm (Noguchi et al., 1999). During moderately dry periods, subsurface flow took place mainly in the soil matrix. Under wet conditions, the short macropore segments link spatially with other preferential flow pathways (Sidle et al., 2001). In British Columbia, Kim et al. (2005) found that, during wet conditions, the lateral outflow of the organic horizon was 400 times higher than during dry conditions.

In both years, greater masses of the less sorbing chemicals (chloride and methomyl) were found in the wick outflow than of the more strongly sorbing chemicals. Thus, matric and preferential transport were affected by the sorption properties of the solutes. However, even the strongly sorbing chlorothalonil was found in the wick outflow of the deeper layers.

Pore Network
The facts that more water leached at the footslope than at the midslope and that the variability of collected water volumes in the wicks of each depth was lower at the footslope indicate that the water-filled network of pores expands downslope. Although at the midslope position only few pores contributed to the wick outflow, at the footslope all wicks were connected to water-bearing pores. At both slope positions, short-distance solute transport mainly occurred in the first soil depth where we observed smaller pores (worm or root holes). Methomyl is the only chemical applied 150 cm away from the profile that was found mainly in the third soil depth. A possible explanation is that methomyl may have been rapidly degraded in the topsoil. Only a small fraction was transported by preferential flow to deeper soil layers and leached to the wick samplers. Long-distance transport occurred mainly in the third depth, where we observed the big macropores during installation of the experiment. This indicates that different pore types contribute to solute transport. The different pore types might be intensively interconnected and activated or deactivated. The linkage can occur through porous zones of organic matter, exchange with shallow bedrock fractures, or through contact with groundwater (Sidle et al., 2000). The hypothesis of at least two water-bearing pore types explains the change of flow depth with varying water suction. Kung et al. (2000) reported a shift toward larger pores with increasing soil wetness. During dryer periods and small rain events, only the smaller pores in the top soil deliver water, but with a wetter soil and stronger rain events, the larger pores in the subsoil react.


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
REFERENCES
 
To investigate the mechanisms of interflow and lateral pesticide transport, two micro-trench experiments were conducted at a hillslope in Northern Thailand. Our data show that wick outflow was not evenly distributed along the soil profile but was concentrated at depths of 0 to 30 cm and 60 to 90 cm. During wet periods, the main interflow took place in the deeper soil. When the average water suction increased to above 40 cm, the flow switched to the topsoil. Slope position, rain amount, and especially antecedent soil moisture affected the amount of interflow. Rain intensity did not have a significant impact on lateral transport processes. At the footslope, the interflow was seven times higher than at midslope position. Moreover, the network of water-bearing pores was more continuous at the footslope. The rain amount needed to induce interflow was 20 mm in both positions.

Lateral solute transport occurred exclusively along preferential pathways and not by matric flow, whereas bulk solute transport was mainly vertical. Short-distance transport occurred in the topsoil, and long-distance transport occurred in the subsoil. Lateral preferential transport of pesticides was observed, but the recovered fraction was negligible. Our results indicate that different pore types contribute to water flow and solute transport in the soil studied: smaller and less continuous pores in the topsoil and larger pores with a higher connectivity in the subsoil. Those systems are activated depending on water suction.


   ACKNOWLEDGMENTS
 
The authors thank Wipawadee Saepueng, Wipa Putha, and Rita Hierl for assistance; Klaus Spohrer for providing soil data and Figure 1; and Ulrich Schuler and Cindy Hugenschmidt for data of the areal density of macropores. This research was funded by Deutsche Forschungsgemeinschaft (German Research Foundation) as part of the SFB 564.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Kahl, G.
Right arrow Articles by Streck, T.
Related Collections
Right arrow Preferential Flow
Right arrow Pesticides
Right arrow Organic Compounds
Right arrow Other Environmental Contamination

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome