Mobility of dichlorprop in the soil-water system as a function of different environmental factors. II. A lysimeter experiment

The Science of the Total Environment, 123/124 (1992) 411-420 Elsevier Science Publishers B.V., Amsterdam

411

Mobility of dichlorprop in the soil-water system as a function of different environmental factors. II. A lysimeter experiment G. Riise a'b, O.M. Eklo b, O. Lode b and M. N a n d r u p Pettersen ~ alsotope and Electron Microscopy Laboratory, Agricultural University of Norway, Box 26, N1432 .~s-NLH, Norway bNorwegian Plant Protection Institute, Box 70, N-1432 ~s NLH, Norway CDepartment of Soil Science, Agricultural University of Norway, Box 28, N-1432 .~s NLH, Norway

ABSTRACT Lysimeter experiments were conducted to study the leaching of [14C]dichlorprop and [3H]water through different soil columns. Results from soil columns collected at Haslemoen showed that an upper silt loam layer (0-18 cm) had higher retention capacity than the underlying sandy loam (35-60 cm) and fine sand (70-95 cm) layers. Leaching through silt clay loam (0-18 cm) columns from Ullensaker was very fast. When the columns received an amount of water corresponding to = 10 mm precipitation, up to 50% of the added [14C]dichlorprop leached through the columns. This is probably due to rapid downward movements through macropores (e.g. cracks). Key words: dichlorprop; mobility; macropores; intact soil columns; lysimeter experiment

INTRODUCTION Phenoxy acids are frequently used herbicides. In 1990, 113 840 kg active ingredient (a.i.) of dichlorprop was sold in Norway. This amount represents 12% of the total herbicides used in 1990. According to a survey carried out in 1987, dichlorprop was one of the pesticides most frequently found in Norwegian surface waters ( G E F O and SPV, 1987). This is in agreement with a Swedish study, in which 259 stream samples were analysed for 80 pesticides. The most frequently found pesticides were dichlorprop and M C P A (Kreiiger and Brink, 1988).

412

o. RIISE ET AL.

The ability of soil to retain and thereby prevent pesticides from reaching' non-target sites depends on geochemical composition and soil texture. In this work the movement of [laC]dichlorprop and [3H]water is studied in lysimeter experiments. Intact soil columns containing soils with different texture are included in the experiments. The results are compared with batch experiments in which the same soil types and solutions are used as in the lysimeter experiments. EXPERIMENTAL

Samples Intact soil columns were taken by pressing PVC-tubes (10 cm i.d.) into the soil. Soil samples were taken from 3 different depths at Haslemoen and from the top layer at Ullensaker. Three parallels were taken from each soil type. The soils varied with respect to particle size distribution, organic carbon (0.1-2.1%), cation exchange capacity (2-18 mequiv./100 g), pH (5.4-6.2) and base saturation (49-92%) (Table 1). The different soil layers from Haslemoen may be classified as silt loam (0-18 cm), sandy loam (35-60 cm) and fine sand (70-95 cm) and the top layer (0-18 cm) from Ullensaker as silty clay loam (Soil Science Society of America, 1979).

TABLE 1 Physico and chemical properties of soil types used in lysimeter experiments Property

Haslemoen

Ullensaker

(0-18 cm)

0-18 cm

35-60 cm

70-95 cm

Size distribution (%) 2-0.6 mm 0.6-0.2 mm 0.2-0.06 mm 0.06-0.02 mm 0.02-0.006 mm 0.006-0.002 mm < 0.002 mm

0 1 14 38 31 8 7

0 0 55 34 7 1 2

0 5 92 3 0 0 0

1 2 4 14 28 29 29

Total C (%) CEC (mequiv./100 g) pH (H20) Base saturation (%)

2.1 12 6.0 49

1.0 4 5.7 85

0.1 2 5.4 92

1.5 18 6.2 54

MOBILITY OF DICHLORPROP IN THE SOIL-WATER SYSTEM: A LYSIMETER EXPERIMENT

413

Pesticides 14C-ring-labelled dichlorprop [2-(2,4-di-chlorophenoxy)propionic acid] with a specific activity of 55.01 /~Ci/mg was used in all the experiments (Agrolinz, 1989). The radiochemical purity was >98%. 14C and 3H measurements

14C and 3H were measured with a Packard Tri-Carb 4530 liquid scintillation counter. The ratio between sample and scintillation cocktail was 1:10. Lysimeter experiment

A solution (25-ml) containing 20 mg (a.i.) inactive dichlorprop, 2.3 #Ci [14C]dichlorprop and 7.5 /~Ci [3H]water/25 ml was applied to the top of every soil column. Inactive dichlorprop was added in the form of a commercial product, Hedonal (Bayer Leverkusen), in which dichlorprop was formulated as K-salt. The feeding of water was controlled by a water programme. The soil columns received 630 ml artificial rainwater/week, corresponding to ~ 11.5 mm precipitation/day. The rainwater had a similar composition as precipitation at Narbuvollen - - a reference site in the project: 'Acid precipitation - - effects on forest and fish' (Overrein et al., 1980). The columns were watered for 4 weeks. At that time = 2500 ml water was eluted from the columns, corresponding to ~ 380 ml precipitation. Water leaching through the columns was sampled once a day. ~4C and 3H were measured in 1-ml x 3 aliquots of the water. Batch experiment

Batch experiments were carried out according to the method given in Riise and Salbu (this volume). The same soil types and pesticide solution as in the leaching experiment were used. Linear and Freundlich isotherms were determined from eight different concentrations of [14C]dichlorprop according to the formulae: C~ = a + KdCw (Linear)

log Cs = log Kf + (l/n) log Cw (Freundlich) where Cs is the amount sorbed (#g/g), Cw is the final concentration in the solution (tzg/ml) and Kd and Kf are the Linear and the Freundlich isotherm constants, respectively.

414

G. RIISE ET AL.

RESULTS A N D DISCUSSION

Lysimeter experiment Haslemoen The leachate curves (Figs 1-3), each based on the average of three columns, show that the movement of dichlorprop through the soil columns varies for the different soil layers. Variation in size and shape of pores may result in dispersion of the waterflow and thereby a broadening of the curve, which is seen (Fig. 1) for the silt loam top layer (0-18 cm). For the sandy loam (35-60 cm) and the fine sand (70-95 cm) the peaks are more distinct and narrow (Figs 2 and 3). This indicates that the front moves downward by piston displacement (small spreading). There is a small delay in movement of [14C]dichlorprop compared to [3H]water (Table 2 and Figs 1-3) in the different soil layers from Haslemoen. The top layer has a higher retention capacity than the underlying layers. Approximately 45% of [~4C]dichlorprop leached through the top layers, while 85% and 92% leached through the middle and bottom layers respectively. This may be due to a broader range of pore sizes (texture) and a higher organic content in the

Haslemoen 0.3

t

i

(0-18 ,

cm) I

...... -'--:

:

I

3 14H C

0.2

0 0 I-(.2 C) 0.1

,"

"%

: J 0.0

. ..:'I

0

" i

500

"~.'.., I

!

"'~'.'~f...~

1000 1500 2000 Leaching (ml)

....... i:

2500

.3000

Fig. 1. Relative percentage amount of 14C and 3H versus leaching (ml) in silt loam columns, n=3.

MOBILITY OF DICHLORPROP IN THE SOIL-WATER SYSTEM: A LYSIMETER EXPERIM

(35-60

Haslemoen 0.3

,

,

,

0 0

cm)

,

l

415

,

3 ...... : t4H -'--: C

0.2 .\

C) ?I .1 -.

Q9 0.1

L ". ~t

:1

'.

"1

":

:I

0.0

•_

0

":

• .'o

I

i

500

"~.

..... F .....

1000 1500 2000 Leaching (ml)

Fig. 2. Relative percentage amount of unms, n = 3.

Haslemoen 0.3

!

!

14C

3000

and 3H versus leaching (ml) in sandy loam col-

(70-g5

I

J 2500

cm)

I

5

O O

0.2

. . . . . . : 141-1 -'-- : C

f,

I I. t

I. H-

t|

(.9 C.)

". i

:1 '

"1 .-'j

0.~

I • °

\l

"t -t "t -%

0.0

I 500

I\'.'~.- ,. £

1000 1500 2000 L e e c h i n g (ml)

I

2500

3000

Fig. 3. Relative percentage amount of t4C and 3H versus leaching (ml) in fine sand columns, n=3.

416

G. RIISE ET AL.

TABLE 2 Results from the lysimeter experiments (n = 3) Soil type

Haslemoen

Ullensaker

(0-18 cm)

0-18 cm

35-60 cm

70-95 cm

760 ± 87 838 ± 156

931 ± 123 978 ± 172

922 ± 48 1044± 65

855 ± 48 919 ± 157

893 ± 84 944 ± 119

957 ± 37 1018 ± 33

Volume (ml) 3H-peak lac-peak Volume (ml) 3H ~(C/CTff2 a 14C ~(C/CTff2

187 ± 144 95 ± 30

Leaching (%) 3H~(C/CT) 100 14C~(C/CT) 100

86± 45±

3 8

89± 85±

2 3

87± 92±

2 3

77± 59±

7 10

aCT = total amount applied (DPM).

upper layer compared to the underlying layers. The results indicate that the mobility of dichlorprop will increase once it has passed the top layer. Less than 100% of [3H]water is found in the leachate. Evaporation and/or isotope exchange with non-mobile water are possible explainations of this result.

Ullensaker The pattern of movement of 14C and 3H in the soil columns from Ullensaker was very different from Haslemoen (Fig. 4). More than 50% of dichlorprop has moved through the columns within 100 ml leachate. The only explanation for this behaviour is a rapid downward movement through macropores (e.g. cracks). The three columns from this site show great variations (Table 2), which indicates non-homogenous distribution of the macropores. Batch experiments Data from the batch experiments, carried out using the same soil types and solutions, as in the lysimeter experiments, fit into the linear as well as the Freundlich sorption isotherm (Table 3). Although relatively high concentrations of dichlorprop (20 ppm) are used in the experiments, maximum sorption levels are not reached (Figs 5-8).

417

MOBILITY OF DICHLORPROP IN THE SOIL-WATER SYSTEM: A LYSIMETER EXPERIMENT

Ullensaker ( 0 - 18cm) 0.8

!ii

0.6

I

:4

~

......

:

3H

---:

c

0.4

0.2

0.0

0

500

1000 i 500 2000 Leaching (rnl)

2500

3000

Fig. 4. Relative percentage amount of t4C and 3H versus leaching (ml) in silty clay columns, n -.~ 3,

TABLE 3 Linear and Freundlich sorption isotherms based on batch equilibrium data of dichlorprop and soil types used in lysimeter experiment Soil type

Haslemoen

Ullensaker

(0-18 cm) Number of observations (different conccentratins) Linear isotherm constant (a) Kd r2 Freundlich isotherm Kr 1/n

r2

0-18 cm

35-60 cm

70-95 cm

8

8

8

8

1.05 1.16 0.993

0.26 0.58 0.990

-0.24 0.46 0.958

1.22 1.00 0.960

1.79 0.89 0.992

0.88 0.83 0.963

0.51 0.84 0.883

1.66 0.88 0.991

418

o. RIISE El" AL

25

~

0 - 1 8 cm 3 5 - 6 0 cm 70-95

:

20

r O1 ~

DICHLORPROP Hoslemoen , u

,

0

15

CP

aL (o

0

10

v

0 5

O--

0

5

z

i

i

10

15

20

25

Cw(~g/~O Fig. 5. Linear isotherms at 17°C.

DICHLORPROP Hc]slemoen

102

i

i

: 0 - 1 8 cm 3 5 - 6 0 cm

101

i

/ 0 ~ all

"~I0 °

10 -I

10 -2

10-2

I

i

I

10-1

100

101

c,(,~g/~=) Fig. 6. Freundlich isotherms at 17°C.

10 2

M O B I L I T Y O F D I C H L O R P R O P IN T H E S O I L - W A T E R S Y S T E M : A L Y S I M E T E R E X P E R I M E N T

DICHLORPROP Ullensoker ( 0 - 1 8 cm) 20

,

,

,

0

15

t3~

o~I0 v

o

ol

0

0 0

i 5

i 10

i 15

20

C ,(/~g,/ml)

Fig. 7. Linear isotherms at I T°C.

DICHLORPROP Ullensaker ( 0 - 1 8 10 2

I

cm)

i

0 10'

ET~ 10 0

10 - I

10 -2 10-2

I 10 -1

I 100

c (~g/~J) Fig. 8. Freundlich isotherms at 17°C.

I 10 I

10 2

419

420

G. RIISE ET AL.

The upper silt loam layer (0-18 cm) from Haslemoen shows a greater sorption capacity than the underlying layers (Figs. 5 and 6). This is in agreement with the results from the lysimeter experiments. If the preferential flow goes through macropores, as supposed in the soil from Ullensaker, only a small part of the solute will get in contact with the soil matrix. In this case the sorption capacity estimated from batch equilibrium may be greatly reduced. However, even though there was a rapid downward movement of the water flow, a significant amount of [laC]dichlorprop is retained (40 4- 10%) in the columns (Table 2). CONCLUSION

According to lysimeter experiments carried out using intact soil columns, different soil layers behaved differently with respect to movement of dichlorprop. The upper silt loam layer (0-18 cm) showed higher retention capacity than the underlying sandy loam (35-60 cm) and fine sand (70-95 cm) layers (texture, org. C). Special hydrological features (e.g., macropores) may have a dominant influence on the mobility of dichlorprop and be the main responsible factor for transport of dichlorprop to non-target sites. ACKNOWLEDGEMENT

We want to thank BASF, DK. for kindly providing ~4C-labelled dichlorprop and our colleagues at the Isotope laboratory for helping with the measurements of the samples. REFERENCES Agrolinz, 1979. Information given by the producer. GEFO and SPV, 1987. Pesticides in surface waters and ground waters. Report from Centre of Soil and Environmental Research and Norwegian Plant Protection Institute, 39 pp. (In Norwegian). Kreiiger, J. and N. Brink, 1988. Losses of pesticides from arable land. Weed and Plant protection Conferences 1988. Swedish University of Agricultural Sciences, Research Information Centre, Jordbruk, 49: 50-61. Overrein, L.N., H.M. Seip and A. Tollan, 1980. Acid precipitation - - effects on forest and fish. Final report of the SNSF project, 1972-1980, 175 pp. Soil Science Society of America, 1979. Glossary of Soil Science Therms. Soil Sci. Soc. Am., Madison, WI.