Pedosphere 18(4): 524532, 2008 ISSN 1002-0160/CN 32-1315/P c 2008 Soil Science Society of China Published by Elsevier Limited

and Science Press

Boron and Zinc Transport Through Intact Columns of Calcareous Soils

M. MAHMOOD-UL-HASSAN, M. S. AKHTAR and G. NABI
Institute of Natural Resources and Environmental Sciences, National Agricultural Research Center, Islamabad 45500 (Pakistan). E-mail: mmh@comsats.net.pk (Received October 15, 2007; revised May 16, 2008)

ABSTRACT
Leaching of boron (B) and zinc (Zn) can be signicant in some pedomorphic conditions, which can cause contamination of shallow groundwater and economic losses. Boron and Zn adsorption and transport was studied using 8.4 cm diameter 28 cm long intact columns from two calcareous soil series with diering clay contents and vadose zone structures: Lyallpur soil series, clay loam (ne-silty, mixed, hyperthermic Ustalc Haplargid), and Sultanpur soil series, sandy loam (coarse-silty, mixed, hyperthermic Ustollic Camborthid). The adsorption isotherms were developed by equilibrating soil with 0.01 mol L1 CaCl2 aqueous solution containing varying amounts of B and Zn and were tted to the Langmuir equation. The B and Zn breakthrough curves were tted to the two-domain convective-dispersive equation. At the end of the leaching experiment, 0.11 L 10 g L1 blue dye solution was also applied to each column to mark the ow paths. The Lyallpur soil columns had a slightly greater adsorption partition coecient both for B and Zn than the Sultanpur soil columns. In the Lyallpur soil columns, B arrival was immediate but the peak concentration ratio (the concentration in solution at equilibrium/concentration applied) was lower than that in the Sultanpur soil columns. The breakthrough of B in the Sultanpur soil columns occurred after about 10 cm of cumulative drainage in both the columns; the rise in euent concentration was fast and the peak concentration ratio was almost 1. Zinc leaching through the soil columns was very limited as only one column from the Lyallpur soil series showed Zn breakthrough in the euent where the peak concentration ratio was only 0.05. This study demonstrates the eect of soil structure on B transport and has implications for the nutrient management in eld soils. Key Words: adsorption isotherm, boron, intact soil column, transport parameters, zinc

Citation: Mahmood-ul-Hassan, M., Akhtar, M. S. and Nabi, G. 2008. Boron and zinc transport through intact columns of calcareous soils. Pedosphere. 18(4): 524532.

INTRODUCTION Boron in soil and water poses ecological and environmental problems and losses of agricultural productivity in the soils where preferential ow is signicant. Kang et al. (2002) ascribed B deciency in cabbage to leaching in humid areas. Surface-applied chemicals have been shown to travel rapidly through small fractions of soil porosity (Radulovich et al., 1992). Simulation models used for predicting water and solute movement often fail to predict the arrival time accurately (Steenhuis et al., 1994; Hateld et al., 1997; Wildenschild et al., 1994; Vervoort et al., 1999). The preferential ow pathways are only a small fraction of the total porosity causing a rapid and accelerated breakthrough of both adsorbing and non-adsorbing solutes (Radulovich et al., 1992; Butcher et al., 1995; Gupta et al., 1999). Accurate estimation of the solute velocities is essential for the prediction of subsoil and groundwater contamination. Solute transport parameters can be predicted accurately only after the breakthroughs over a range of ow rates are known. Studies on B transport are limited in calcareous soils. Boron adsorption is considered to be reversible but the use of models based on the assumption of local equilibrium often provides poor descriptions of B transport in soil columns. Also, B transport is strongly related to soil pH (Communar and Keren, 2005). It appears that soil pH and occurrence of macropores are two important characteristics controlling the

BORON AND ZINC TRANSPORT THROUGH SOIL COLUMNS

525

leaching of B. Similarly, the leaching of Zn, which aects its fraction distribution in soils and availability to plants, depends on the soil physical properties (Alvarez et al., 2001; Voegelin et al., 2003). Adsorption coecient (k) by batch technique helps to explain the results of leaching studies, and given the pore water velocity, the leaching experiments can also yield k (Schweich et al., 1983; Bond and Phillips, 1990a; Barnett et al., 2000). The batch experiment has the disadvantages of breakdown of soil aggregates during sample agitation, the soil/solution ratio being relatively small, and the dierences in mass transfer and hydrodynamic conditions, which cumulatively result in inappropriate estimates of the degree of adsorption (Schweich et al., 1983). The column technique, where the solution of the convectiondispersion equation (CDE) helps to obtain k by tting breakthrough curves (BTC), overcomes some of these limitations but the disadvantage of the column technique is the absence of an accurate solution for a nonlinear CDE with equilibrium adsorption (Veldhuizen et al., 1995). The disadvantage of column study can be partly overcome by the application of semi-analytical solutions of a nonlinear transport CDE (Bond and Phillips, 1990b). Boron application is widely recommended for eld crops (Rashid et al., 2002) but little information is available on B adsorption and transport in calcareous soils. As part of a study on leaching of fertilizer nutrients, objectives of this study were to determine whether the preferential ow of B and Zn actually occurs in calcareous soils and to determine the transport parameters for these fertilizer nutrients. MATERIALS AND METHODS This study was carried out at the Institute of Natural Resources and Environmental Sciences, National Agricultural Research Center, Islamabad, using intact columns of two calcareous soils, Lyallpur and Sultanpur soil series (73 10 73 45 E, 31 10 31 30 N) from Bari Doab, an alluvial landmass between Ravi and Sutloj rivers with semi-arid subtropical continental climate (Mian, 1968). The Lyallpur soil series (pH 8.2), clay loam (ne-silty, mixed, hyperthermic Ustalc Haplargid) developed on late Pleistocene, and Sultanpur soil series (pH 8.0), sandy loam (coarse-silty, mixed, hyperthermic Ustollic Camborthid) developed on sub-recent level oodplain, are the most dominant soils in the region used for maize-potato-maize cropping sequence. The saturated hydraulic conductivities, bulk densities, calculated mean pore volumes, and total porosities of the Sultanpur and Lyallpur soil series were 125 and 152 mm d1 , 1.49 and 1.42 Mg m3 , and 0.43 and 0.46 m3 m3 , respectively. The soil columns, 8.4 cm diameter and 28.0 cm long, were excavated by inserting PVC pipes, and brought to laboratory. The bottom and surface of the columns were leveled by removing 1.5 cm soil layer from each side. The bottom soil layer was supported by small gravels and plastic netting xed on a support PVC pipe of relatively large diameter resting on a ceramic funnel. The column was saturated from the bottom by 0.01 mol L1 CaCl2 electrolyte solution. For the leaching experiment, 10 mg L1 solution containing ZnSO4 and H3 BO3 prepared in 0.01 mol L1 CaCl2 was applied on the column surface at a constant ow rate controlled by 2 cm water head. The leachate was collected in plastic containers at time intervals and analyzed for B and Zn. Boron and Zn adsorption coecients were determined from adsorption isotherms developed through batch experiments (Syers et al., 1973; Castro and Rolston, 1977). The surface soil from each series was passed through a 2-mm sieve and 5 g soil was equilibrated in triplicate with 30 mL 0.01 mol L1 CaCl2 aqueous solution containing 0, 8, 20, 40, 80, 120, 160, 240, 320, 400, 480 mg B L1 as H3 BO3 . The suspension was shaken overnight on an end to end shaker at room temperature and then centrifuged. The pH of the supernatant was recorded and B in the supernatant was analyzed. Separately, 5 g soil in 30 mL of 0.01 mol L1 CaCl2 with 0, 25, 50, 100, 200, 300, 400, 500, 600, 700, and 800 mg Zn was shaken overnight and the supernatant was analyzed for Zn. The amount of B or Zn added but not recovered in solution was assumed to be adsorbed. Adsorption was expressed on a dry soil weight basis. In addition to the leaching experiment, the ow paths were marked by applying brilliant blue dye to the columns. 0.11 L 10 g L1 FD&C blue dye No. 1 solution was applied on the column surface without causing runo (Flury et al., 1994). After 24 hours, each column was cut perpendicularly with

526

M. MAHMOOD-UL-HASSAN et al.

3 cm increments towards the center and the faces were photographed successively. Boron was analyzed calorimetrically using the azomethine-H and the absorbance was recorded at 420 nm (Keren, 1996). Zinc was analyzed using an atomic absorption spectrophotometer. The adsorption isotherms as determined from the batch experiment were tted to the Langmuir equation (Castro and Rolston, 1977), which has the form: X= bKC 1 + KC (1)

or is rearranged in linear form: 1 C C = + X Kb b (2)

where X equals the adsorption by soil (mg kg1 ), C is the concentration in solution at equilibrium (mg L1 ), K is a constant related to the binding strength (L mg1 ), and b is the maximum adsorption on the soil (mg kg1 ). When C = 1/K, then X = b/2, or half of the sites on the soil are lled with adsorbate. Boron and Zn adsorption isotherms were plotted as C/X versus C and regressed linearly. The reciprocal of the slope of the regression line yielded b, the maximum B (or Zn) adsorption, and the Langmuir constant K was obtained by dividing the slope by the intercept. To investigate the degree of adsorption at the application concentration of 10 mg L1 , we determined the adsorption partition coecient, k, by nding the tangent to the Langmuir isotherm at this concentration (Akhtar et al., 2003). Dierentiating Eq. 1 with respect to C gives: X 1 bK 2 C = + bK C 1 + KC (1 + KC)2 or k= 1 bK 2 C + bK 1 + KC (1 + KC)2 (4) (3)

The CDE model for one-dimensional transport of reactive solutes subject to adsorption in one or two domains has been solved for several boundary conditions given in Parker and van Genuchten (1984) and other textbooks: R C 2C C v =D t X 2 X (5)

where R is the retardation factor for adsorbing solute, D is the dispersion coecient, t is the time, and v is the velocity of the solute and equals q/((k + ), where q is the water ux density, is the mobile water fraction (Parker and van Genuchten, 1984), is the bulk density (Mg m3 ), and is the total porosity (m3 m3 ). To solve the dierential equation, a constant adsorption partition coecient is employed, resulting in solutions in which the movement of solutes can be scaled with a retardation coecient of the form (Toride et al., 1995): R=1+ k (6)

All the breakthrough curves were tted to CDE by partitioning into mobile/immobile fractions using STANMOD, a Windows-based computer code, which uses analytical solutions of deterministic non-equilibrium CDE (Simunek et al., 1999; Toride et al., 1995). The input data consisted of solute concentration ratio C/C0 , where C0 is the concentration applied, and drainage depth (cm) as proxy for the solute concentration and time, respectively. The boundary conditions included a characteristic length of 0.28 m and the total pulse input equaling to the drainage depth of the application phase. The

BORON AND ZINC TRANSPORT THROUGH SOIL COLUMNS

527

R values calculated using Eq. 6 were used. No decay or production terms were included. Furthermore, the uniqueness of the tted parameters was veried by STANMOD run repeatedly for each experiment with varying initial values of the input parameters (Langer et al., 1999). The tted parameters from the analytical solution were accepted once the goodness of t (r2 ) was the highest and the t value for the individual parameter was at least one (Kim et al., 2005). The mean square of error for the model was < 0.01 in most cases. The tted ow parameters for the breakthrough of B and Zn were: Vm , the velocity in the mobile region dened as m v/ (cm h1 ), where m is the mobile water fraction (m3 m3 ); Dm , the dispersion coecient in the mobile region dened as m D/ (cm2 h1 ); , the mobile water partitioning coecient ( = m /); and , the mass transfer function (Simunek et al., 1999; Toride et al., 1995). RESULTS AND DISCUSSION Flow path characteristics The columns from both soils showed dierences in ow paths as marked by the blue dye. Preferential ow of blue dye was visible after uniform wetting of the surface 2 cm column depth but the nature of ow was dierent in each soil. In the Sultanpur soil series, a massive loam soil, blue dye after uniformly wetting surface 2 cm soil owed through 3 cm wide ngers and had no association with any pedological feature (Fig. 1a, b), indicating existence of nger ow in this loamy soil. In the Lyallpur soil series, the blue dye moved through macropores and then moved outward perpendicular to the ow paths, suggesting diusion into the matrix (Fig. 1c, d). The distribution of dyed area reected dierences in the nature of ow paths in the soils: ngering and, possibly, macropore ow in the massive loam Sultanpur soil series and only macropore ow in the Lyallpur soil series. Instability in the wetting front causes nger ow and it occurs in homogenous soils (Wang et al., 2003), while macropore ow is associated with biopores (earthworm burrows, decayed root channels, etc.) and the interpedal voids and ssures form owing to shrinking and swelling (Beven and Germann, 1982).

Fig. 1 Patterns of blue dye staining in two columns from two soil series: Sultanpur, a sandy loam (a and b) and Lyallpur, a clay loam (c and d).

528

M. MAHMOOD-UL-HASSAN et al.

Boron adsorption Boron adsorption isotherms for both soils are depicted in Fig. 2a. Initially, there was a rapid increase in B adsorption associated with the adsorption on the solid surface in both the soils. Then, it was followed by a slow rise of isotherm associated with the diusion of B into the clay mineral. Boron adsorption isotherm had a steeper slope for the Lyallpur compared to the Sultanpur soil series. Previously, Couch and Grim (1968) proposed a two-step mechanism for B retention in soils. Grin and Burau (1974) and Sharma et al. (1989) observed a slow B desorption from clay soil and assumed that it could be due to a slow B diusion from the interior surfaces of clay mineral to the solution phase.

Fig. 2 Boron adsorption isotherms (mean of three replicates) for Lyallpur and Sultanpur soil series (a) and the Langmuir equation t of the B adsorption isotherms for the Lyallpur (b) and Sultanpur (c) soil series. C is the concentration in solution at equilibrium and X is the adsorption by soil.

The B adsorption isotherms tted the Langmuir equation (Eq. 2) well (Fig. 2b, c) (r2 = 0.75 in the Lyallpur and 0.81 in the Sultanpur soil series). The maximum B adsorption and the binding strength determined from the reciprocals of the slope and the intercept, respectively, of the regression line of Langmuir equation were greater in Lyallpur compared with the Sultanpur soil series (Table I). This dierence appears to be related to the clay content as previously reported (Communar and Keren, 2005).
TABLE I Langmuir adsorption parametersa) calculated from the linear B and Zn adsorption isotherms Soil Boron Zinc Lyallpur Sultanpur Lyallpur Sultanpur r2 0.75 0.81 0.88 0.85 Intercept 0.081 0.102 0.338 0.044 Slope 0.069 0.070 0.010 0.010 b mg kg1 14.43 14.25 104 94 K L mg1 0.86 0.69 0.25 0.23 k 1.17 1.10 20.80 18.00 R 4.10 3.90 65.10 63.37

a) r 2 is the regression coecient; intercept is 1/Kb; slope of the regression line between C/X vs. C is 1/b, where C is the concentration in solution at equilibrium (mg L1 ) and X is the adsorption by soil (mg kg1 ); b is the maximum adsorption; K is a constant relating to the binding strength; k is the adsorption partition coecient tangent to isotherm at the concentration applied (10 mg L1 ); and R is the retardation coecient.

Adsorption partition coecients at C0 = 10 mg L1 , the concentration applied during the leaching experiment calculated with Eq. 4 using the tted Langmuir parameters, are given in Table I. The adsorption partition coecient is assumed to be constant over the range of concentrations that occurred in the soil. The surface soils had the adsorption partition coecients in the range of 1.17 in the Lyallpur and 1.1 in the Sultanpur soil series (Table I). Similarly, the retardation coecient calculated from Eq. 6 was slightly greater in the Lyallpur as compared to the Sultanpur soil series (Table I). Therefore, the Lyallpur soil series had slightly greater B adsorption potential than the Sultanpur series, which was

BORON AND ZINC TRANSPORT THROUGH SOIL COLUMNS

529

related to their dierence in clay content though they had very similar pH. Zinc adsorption Zinc adsorption isotherms for the soils are shown in Fig. 3a. Initially, there was a rapid increase in Zn adsorption associated with the adsorption on the solids surface, which was then followed by a slow rise of isotherm associated with diusion into the clay mineral (Couch and Grim, 1968; Grin and Burau, 1974; Sharma et al., 1989). Both soils had similar adsorption trend at low equilibrium concentration ( 1 mg L1 ), while at greater equilibrium concentrations, the Lyallpur soil series adsorbed relatively greater Zn than the Sultanpur soil series. For example, the Lyallpur soil series adsorbed about 80 mg kg1 and the Sultanpur soil series adsorbed approximately 70 mg kg1 at the equilibrium concentration of 11 mg L1 and there was no further increase in adsorption with increase in equilibrium concentration after 12 mg L1 .

Fig. 3 Zinc adsorption isotherms (mean of three replicates) for the Lyallpur and Sultanpur soil series (a) and the Langmuir equation t of the Zn adsorption isotherms for the Lyallpur (b) and Sultanpur (c) soil series. C is the concentration in solution at equilibrium and X is the adsorption by soil.

The isotherms tted to the Langmuir equation are depicted in Fig. 3b, c for both soils. The Langmuir equation tted the data well (r2 = 0.88 in the Lyallpur and 0.85 in Sultanpur soil series). The Lyallpur soil series had greater maximum Zn adsorption capacity (b) and binding strength (K) than the Sultanpur soil series (Table I). The adsorption partition coecient (k) calculated with Eq. 4 using the Langmuir parameters (Table I) at C0 = 10 mg L1 , which is the tangent to the slope at this concentration, was 21 in the Lyallpur and 18 in the Sultanpur soil series (Table I). Therefore, the Lyallpur soil series had slightly greater Zn adsorption capacity than the Sultanpur soil series and it appeared that this may be related to the clay content. Boron breakthrough Breakthrough curves plotted as a function of time, pore volumes of euent, or cumulative drainage, are a common tool to determine the transport properties (Jardine et al., 1988; Vereecken et al., 1990; Zurmhl, 1998). Boron breakthrough in Sultanpur soil columns occurred after 10 cm cumulative drainage and the relative concentration (C/C0 ) peaked to 0.9 after 32 cm of cumulative drainage in Column 1 and to 1.0 after 40 cm of cumulative drainage in Column 2 (Fig. 4a). In both soils, the B breakthrough occurred earlier than one pore volume. In the Lyallpur soil columns, the breakthrough was immediate owing to high preferential ow; but after initially steep slope, C/C0 reached only 0.4 after 24 cm of cumulative drainage in Column 1. Also, this curve shows a long tail (Fig. 4b), implying that the concentration decrease was slow during the ushing phase. Diusion of solutes perpendicular to the ow path into matrix during the application phase and seeping back into the macropore during ushing caused the long tail of BTC. The blue dye pattern in the soil also showed diusion perpendicular to the macropore (Fig. 1c, d). Although in Column 2 of the Lyallpur soil series the B breakthrough was also immediate owing to high preferential ow, the C/C0 reached its peak of 0.1 after 13 cm of cumulative

530

M. MAHMOOD-UL-HASSAN et al.

drainage (Fig. 4b). Therefore, we see delay in terms of concentration but not in terms of arrival time.

Fig. 4 Boron breakthrough curves in two columns from two soil series: Sultanpur, a sandy loam (a) and Lyallpur, a clay loam (b). C is the concentration in solution at equilibrium and C0 is the concentration applied.

In contrast, there was a relatively sharp decline in percolate B in the Sultanpur compared to the Lyallpur soil columns during the ushing phase when B application was stopped and B-free water was started. Macropore ow usually occurs near saturation; it does not depend on the water content of the bulk matrix, but instead depends principally on the surface boundary conditions (Mahmood-ul-Hassan, 1998). Zinc breakthrough Breakthrough of Zn was very limited within the drainage depth in the two soil series. Only one column from the Lyallpur soil series showed C/C0 of 0.05 toward the end of the leaching experiment (data not presented). Transport parameters for Zn were not determined because of lack of breakthrough; however, we had R values calculated from Eq. 6 using the k values from the batch experiment (Table I). Model tting The two-domain CDE tted well to the B breakthrough data except for one column of the Lyallpur soil series, where r2 was only 0.44 (Table II). Boron under 20 mm water head moved 5.3 cm per cm of drainage in the Sultanpur soil columns as determined by tting the breakthrough curves. One column of the Lyallpur soil series was as fast as the columns of the Sultanpur soil series but the other was relatively slow. The water partitioning coecient in the columns of the Sultanpur soil series was 0.97 compared to 0.30 and 0.07 for the two columns of the Lyallpur soil series. This indicated that larger volume fraction of the Sultanpur soil series had taken part in the ow process, implying a lesser degree of preferential ow compared to the Lyallpur soil series, which conformed to blue dye ow patterns, i.e., nger ow in the Sultanpur and macropore ow in the Lyallpur soil series. However, the Lyallpur soil columns had high dispersion in the mobile region (Table II). We believe that the high degree of dispersion was probably due to rapid breakthrough of solutes owing to preferential ow paths.
TABLE II Solute transport parametersa) determined using the convection-dispersion equation Soil Sultanpur Lyallpur
a) r 2

Column 1 2 1 2

r2 0.94 0.98 0.88 0.44

0.97 0.97 0.30 0.07

w 0.51 99.00 0.87 79.44

Vm cm 5.64 4.91 5.28 2.23 h1

Dm cm2 h1 16.76 12.47 42.45 458.66

is the regression coecient; is the water partitioning coecient; is the mass transfer function; Vm is the velocity of the mobile liquid phase; and Dm is the dispersion coecient of the mobile liquid phase.

BORON AND ZINC TRANSPORT THROUGH SOIL COLUMNS

531

CONCLUSIONS We investigated the leaching of B and Zn through two soils, the Sultanpur and Lyallpur soil series, by taking replicated intact columns. We also marked the ow paths in the columns used and determined B and Zn adsorption potential by batch experiment. This study provides a comparison of the observed (from breakthrough curves) and calculated rst arrival time (by tting CDE solute transport model) and shows that both soils had preferential ows that varied in nature and magnitude. The Lyallpur soil, owing to its greater clay content, adsorbed greater B and Zn compared to the Sultanpur soil, but its macropore ow had earlier breakthrough of B. However, the solute velocity in the columns of the Lyallpur soil series was not higher than that of the Sultanpur soil series mainly because the solute movement was attributed to high dispersion. The study demonstrates the eect of soil structure on solute transport and has implications for the nutrient management in eld soils. REFERENCES
Akhtar, M. S., Richards, B. K., Medrano, P. A., deGroot, M. and Steenhuis, T. S. 2003. Dissolved phosphorus from undisturbed soil cores: Related to adsorption strength, ow rate, or soil structure? Soil Sci. Soc. Am. J. 67: 458470. Alvarez, J. M., Novillo, J., Obrador, A. and Lpez-Valdivia, L. M. 2001. Mobility and leachability of zinc in two soils o treated with six organic zinc complexes. J. Agr. and Food Chem. 49: 3 8333 840. Barnett, M. O., Jardine, P. M., Brooks, S. C. and Selim, H. M. 2000. Adsorption and transport of U(VI) in subsurface media. Soil Sci. Soc. Am. J. 64: 908917. Beven, K. and Germann, P. 1982. Macropores and water ow in soils. Water Resour. Res. 18: 1 3111 325. Bond, W. J. and Phillips, I. R. 1990a. Cation exchange isotherms obtained with batch and miscible-displacement techniques. Soil Sci. Soc. Am. J. 54: 722728. Bond, W. J. and Phillips, I. R. 1990b. Approximate solutions for cation transport during unsteady, unsaturated soil water ow. Water Resour. Res. 26: 2 1952 205. Buchter, B., Hinz, C., Flury, M. and Fluhler, H. 1995. Heterogeneous ow and solute transport in an unsaturated stony soil monolith. Soil Sci. Soc. Am. J. 59: 1421. Castro, C. L. and Rolston, D. E. 1977. Organic phosphate transport and hydrolysis in soil: Theoretical and experimental evaluation. Soil Sci. Soc. Am. J. 41: 1 0851 092. Communar, G. and Keren, R. 2005. Equilibrium and nonequilibrium transport of boron in soil. Soil Sci. Soc. Am. J. 69: 311317. Couch, E. L. and Grim, R. E. 1968. Boron xation by illites. Clay Mineral. 16: 249256. Flury, M., Fluhler, H., Jury, W. A. and Leuenberger, J. 1994. Susceptibility of soils to preferential ow of water: A eld study. Water Resour. Res. 30: 1 9451 954. Grin, R. A. and Burau, R. G. 1974. Kinetic and equilibrium studies of boron desorption from soil. Soil Sci. Soc. Am. J. 38: 892897. Gupta, A., Destouni, G. and Jensen, M. B. 1999. Modelling tritium and phosphorous transport by preferential ow in structured soil. J. Contam. Hydrol. 35: 389407. Hateld, K. K., Warner, G. S. and Guillard, K. 1997. Bromide and FD&C blue no 1 dye movement through intact and packed soil columns. Trans. ASAE. 40: 309315. Jardine, P. M., Wilson, G. V. and Luxmoore, R. J. 1988. Modeling the transport of inorganic ions through undisturbed soil columns from two contrasting watersheds. Soil Sci. Soc. Am. J. 52: 1 2521 259. Kang, Y., Tsuji, M., Kosuga, H., Sakurai, K. and van Tran, K. 2002. Evaluating boron behavior in soil and water as a step toward achieving sustainable agriculture. In IUSS (ed.) Proceedings of the 17th World Congress of Soil Science, 1421 August, 2002, Bangkok, Thailand. IUSS, Bangkok. Keren, R. 1996. Boron. In Sparks, D. L. et al. (eds.) Methods of Soil Analysis. Part 3. Chemical and Microbiological Properties. ASA and SSSA, Madison, WI. pp. 603626. Kim, Y.-J., Darnault, C. J. G., Bailey, N. O., Parlange, J.-Y. and Steenhui, T. S. 2005. Equation for describing solute transport in eld soils with preferential ow paths. Soil Sci. Soc. Am. J. 69: 291300. Langer, H. W., Gaber, H. M., Wraith, J. M., Huwe, B. and Inskeep, W. P. 1999. Preferential ow through intact soil cores: Eect of matric head. Soil Sci. Soc. Am. J. 63: 1 5911 598. Mahmood-ul-Hassan, M. 1998. Water movement through unsaturated zone. Ph.D. Thesis, University of Reading, UK. 156pp. Mian, M. A. 1968. Reconnaissance Soil Survey of Sahiwal District. Soil Survey of Pakistan, Lahore. 191pp. Parker, J. C. and van Genuchten, M. T. 1984. Determining transport parameters from laboratory and eld tracer experiments. Va. Agr. Exp. Stat. B. 84(3): 193.

532

M. MAHMOOD-UL-HASSAN et al.

Radulovich, R., Sollin, P., Baveye, P. and Solorzano, E. 1992. Bypass water ow through unsaturated micro aggregated tropical soils. Soil Sci. Soc. Am. J. 56: 721726. Rashid, A., Raq, E. and Ryan, J. 2002. Establishment and management of boron deciency in eld crops in Pakistan. In Goldbach, H. E. et al. (eds.) Boron in Plant and Animal Nutrition. Kluwer Academic/Plenum Publishers, New York. pp. 339348. Schweich, D., Sardin, M. and Gaudet, J. P. 1983. Measurement of a cation exchange isotherm from elution curves obtained in a soil column: Preliminary results. Soil Sci. Soc. Am. J. 47: 3237. Sharma, H. C., Pasricha, N. S. and Bajwa, M. S. 1989. Comparison of mathematical models to describe boron desorption from salt aected soils. Soil Sci. 147: 7784. Simunek, J., Van Genuchten, M. T., Sejna, M., Toride, N. and Leij, F. J. 1999. The STANMOD Computer Software for Evaluating Solute Transport in Porous Media Using Analytical Solution of Convection-Dispersion Equation. U.S. Salinity Laboratory Agri. Res. Service U.S. Dept. Agri., Riverside, California. 20pp. Steenhuis, T. S., Boll, J., Shalit, G., Selker, J. S. and Merwin, I. A. 1994. A simple equation for predicting preferential ow solute concentrations. J. Environ. Qual. 23: 1 0581 064. Syers, J. K., Browman, M. G., Smillie, G. W. and Corey, R. B. 1973. Phosphate sorption by soils evaluated by the Langmuir adsorption equation. Soil Sci. Soc. Am. J. 37: 358363. Toride, N., Leij, F. J. and van Genuchten, M. T. 1995. The CXTFIT Code for Estimating Transport Parameters from Laboratory or Field Tracers Experiments. Version 2.1. Research Report No. 137. U.S. Salinity Laboratory Agri. Res. Service U.S. Dept. Agri., Riverside, California. 121pp. Veldhuizen, M., Appelo, C. A. J. and Grioen J. 1995. The (least squares) quotient algorithm as a rapid tool for obtaining isotherm form from column elution curves. Water Resour. Res. 31: 849857. Vereecken, H., Maes, J. and Feyen, J. 1990. Estimating unsaturated hydraulic conductivity from easily measured soil properties. Soil Sci. 149: 112. Vervoort, R., Radclie, W. D. E. and West, L. T. 1999. Soil structure development and preferential solute ow. Water Resour. Res. 35: 913928. Voegelin, A., Barmettler, K. and Kretzschmar, R. 2003. Heavy metal release from contaminated soils. Soil Sci. Soc. Am. J. 32: 865875. Wang, Z., Tuli, A. and Jury, W. A. 2003. Unstable ow during redistribution in homogeneous soil. Vadose Zone J. 2: 5260. Wildenschild, D., Jensen, K. H., Villholth, K. and Illangasekare, T. H. 1994. A laboratory analysis of the eect of macropores on solute transport. Ground Water. 32: 381389. Zurmhl, T. 1998. Capability of convection-dispersion transport models to predict transient water and solute movement in undisturbed columns. J. Contam. Hydrol. 30: 101128.