Effect of water quality on the leaching of potassium from sandy soil

Effect of water quality on the leaching of potassium from sandy soil

ARTICLE IN PRESS Journal of Arid Environments Journal of Arid Environments 68 (2007) 624–639 www.elsevier.com/locate/jnlabr/yjare Effect of water qu...

234KB Sizes 1 Downloads 24 Views

ARTICLE IN PRESS Journal of Arid Environments

Journal of Arid Environments 68 (2007) 624–639 www.elsevier.com/locate/jnlabr/yjare

Effect of water quality on the leaching of potassium from sandy soil Z. Kolahchi, M. Jalali Department of Soil Science, College of Agriculture, Bu-Ali Sina University, Hamadan, Iran Received 18 October 2005; received in revised form 24 April 2006; accepted 27 June 2006 Available online 24 August 2006

Abstract When potassium (K+) fertilizers are applied to soil, K+ is subject to displacement through the soil profile. More generally, the application of K+ fertilizers to sandy soils with low clay content and small buffer capacity, in which K+ does not interact strongly with the soil matrix, results in localized increases in K+ concentration in the soil solution. Losses of K+ depend on the concentration of calcium (Ca2+) as a competing ion in the leaching water and the amount of water that passes through the soil. In this study, we examined the adsorption and movement of applied K+ in columns of sandy soil. Glass tubes, 4.8 cm in diameter and 40 cm in length, were packed with either native soil or Ca2+-saturated soil. The resulting 10-cm-long column of soil had a bulk density of 1.65 g cm3. Native soil was leached with distilled water and CaCl2 solutions of various concentrations. In the Ca2+-saturated soil, a pulse of K+ was leached with CaCl2 solutions of various concentrations or distilled water. Increasing the CaCl2 concentration from 3 to 15 mM resulted in earlier breakthrough, a higher peak concentration of K+, and greater amounts of leached K+. The breakthrough curve for K+, when leached with distilled water, showed very low concentrations and was more delayed than the other treatments. In Ca2+-saturated soil, the amount of K+ leached increased as Ca2+ concentration increased, with up to 54% of the added pulse K+ being removed from 10 pore volumes (Pv) (387 mm) of 15 mM CaCl2. The presence of Ca2+ in irrigation water and soil minerals able to release Ca2+ is important in determining the amount of K+ leached from soils. Large amounts of K+ are leached from soils in areas where crops are irrigated with water that contains significant concentrations of Ca2+ and other cations. r 2006 Elsevier Ltd. All rights reserved. Keywords: Leaching; Potassium; Native potassium; Water quality; Iran

Corresponding author. Tel.: +98 811 4227090; fax: +98 811 4227012.

E-mail address: [email protected] (M. Jalali). 0140-1963/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jaridenv.2006.06.010

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

625

1. Introduction The fate of the nutrient K+ has received less attention than that of nitrogen or phosphorus. The chemistry and fertility of K+ in arid and semi-arid regions are poorly studied because soils in these regions are generally well supplied with K+; however, soils that are intensively cropped become progressively depleted in plant-available K+. The forms of K+ in soil, in order of their availability for leaching, are solution, exchangeable, non-exchangeable and mineral (Martin and Sparks, 1985; Sparks and Huang, 1985). Exchangeable and solution forms are primarily involved in leaching. The application of K+ fertilizers to most sandy soils with low clay content and small buffer capacity, in which K+ does not interact strongly with the soil matrix, results in localized increases in K+ concentration in the soil solution; subsequently, K+ is leached by rainfall or irrigation water. In arid and semi-arid regions, the leaching of K+ is enhanced by the presence of calcite and gypsum (Jalali and Rowell, 2003). In addition to clay type and content, organic matter content and the amount of applied K+ (Johnston et al., 1993), the leaching of K+ is also dependent on the concentrations of other cations, especially Ca2+ in the soil solution. The source of Ca2+ for displacing K+ is either saline solution (Rowell, 1985) or the weathering of soil minerals (Shainberg et al., 1981), especially those that contain gypsum and calcite. Calcium is the dominant cation in soil water and at the exchange sites, and competes with K+ for exchange sites when K+ fertilizers are applied to the soil. Johnston and Goulding (1992) suggested that approximately 1 kg K+ ha1 was lost for every 100 mm of rainwater leached through the soil in the field, but this value may be larger if K+ is displaced with a solution that contains a higher concentration of Ca2+ ions. Johnston et al. (1993) studied a sandy loam soil and measured 20–80 kg ha1 leaching of K+ from the soil profile over 1.5 years. Heming and Rowell (1997) analysed chalky soils in laboratory studies and measured leaching of 9 and 74 kg K+ ha1 following leaching equivalent to 1 year of throughflow in the field. It is important to note that the leaching of K+ in arid and semi-arid regions is different to that in temperate regions. A characteristic of arid and semi-arid regions is low rainfall and the necessity of irrigation. The shortage of quality water resources is becoming an important issue in arid and semi-arid regions of the world. In these regions, the availability of non-saline river or canal water is limited and prioritized to supplying urban areas (Beltran, 1999). Ground-water is commonly the only source of irrigation, although its quality is usually low because of limited rainfall and high rates of evaporation. Thus, there is an increasing need to irrigate using low- to medium-quality ground-water. Irrigation with water in which the concentrations of Ca2+, Mg2+, and Na+ are higher than those in high-quality water leads to an increase in K+ desorption and leaching (Meiri et al., 1984; Feigenbaum and Meiri, 1988). This K+ may be more readily available to plant roots, but it is also easily leached down beyond the root zone. Feigenbaum (1986) reported losses equivalent to 90–300 kg K+ ha1 when 430 mm solutions containing 5 and 50 cmolc l1 of mixed NaCl–CaCl2 were applied to soil columns in the laboratory. BarTal et al. (1991) showed that irrigation water with high salinity can leach native and applied K+ from the soil. Jalali and Rowell (2003) reported losses equivalent to 29–387 kg ha1 when 780 mm of distilled water was applied to calcite- and gypsum-bearing soil columns in the laboratory. Therefore, an increase in K+ concentration can be expected in ground-water within infiltration areas subjected to agricultural land use. Such increases

ARTICLE IN PRESS 626

Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

can even lead to a breach of the drinking water limit for K+ (12 mg l1) (WHO, 1993; Griffioen, 2001). Salinity is the principal concern for irrigated agriculture in the Hamadan area, western Iran. The salinity hazard for water wells in this area is classified as either medium, high, or very high salinity (Jalali, 2002a). There are many studies on the adverse effects of using saline water on soil (Tanji, 1990; Rhoades et al., 1992; Marwan and Rowell, 1995; Minhas, 1996; Rowell and Dimitrios, 2002), but few experimental studies have examined the leaching of K+ from sandy loam under arid and semi-arid conditions. The objectives of the present study are to investigate the effects of rainwater and poor-quality irrigation water on the leaching of K+. 2. Materials and methods 2.1. Soil The analysed soil sample was taken from the 0 to 30 cm layer of agricultural soil in Hamadan, western Iran. The sample was selected to represent typical sandy loam soil that is continuously cultivated (for grapes (Vitis vinifera L.)). The soil was air-dried and passed through a 2-mm mesh sieve before being stored in polyethylene bags. Soil pH, organic matter, exchangeable K+, cation exchange capacity (CEC), texture and equivalent carbonate calcium were determined according to the methods of Rowell (1994). The soil was a sandy loam of the Azandarian Series (Typic Calcixerolic Xerochrept). Some of the relevant chemical and physical properties of an air-dry o2 mm sample are as follows: pH (H2O) ¼ 7.1; organic matter ¼ 37.1 g kg1; exchangeable K+ ¼ 0.516 cmolc kg1; CEC ¼ 12.6 cmolc kg1; clay ¼ 157 g kg1; sand ¼ 620 g kg1, and equivalent carbonate calcium ¼ 47 g kg1. This soil herein is termed native soil. 2.1.1. Ca2+-saturated soil Ca2+-saturated soil was prepared by leaching the native soil with 30 Pv of 50 mM CaCl2 followed by 30 Pv of 5 mM CaCl2 (Jalali and Rowell, 2003) (1 Pv is the volume of water held at saturation). The soil was then leached with 95% ethanol to remove free salts before being air-dried and passed through a 2-mm sieve. It should be noted that even after extensive leaching with CaCl2 solution, the soil still contained exchangeable K+ (i.e. K+ extractable in NH4OAc) apparently related to the release of K+ from non-exchangeable forms. 2.2. Leaching experiments The leaching columns consisted of Pyrex tubes of 30 cm length and an internal diameter of 4.8 cm. The soil was seated at a height of 10 cm by uniform tapping with a wooden rod to achieve a uniform bulk density of 1.65 g cm3. The texture of the studied soil is sandy; the bulk densities of sandy soils are 1.4–1.7 g cm3 (Rowell, 1994). Packing of the soil into the column gives a bulk density of 1.65 g cm3, which is within the range for sandy soils and is slightly more than the bulk density of the sampled soil in the field (approximately 1.55 g cm3). A Whatman No. 42 filter paper and a piece of nylon mesh were placed at the bottom of the leaching column. Each column was initially saturated with distilled water and then

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

627

leached with either distilled water or CaCl2 solution with varying concentrations (3, 5, 10, and 15 mM). Distilled water is used to represent the relatively low-salinity water sourced from rainfall or snowmelt in arid and semi-arid regions (Rowell, 1994; Al-Wabel et al., 2002). Although distilled water and rain do not necessarily have similar Ca2+ concentrations, when they enter the soil, CaCO3 dissolution is the main source of Ca2+ for displacing other cations from exchange sites. To simulate the long-term leaching of K+, the leaching column was left until more than 75% of applied K+ was leached from the soil via 3–15 mM CaCl2 leaching solutions. In native soil, leaching was continued until more than 90% of exchangeable K+ was leached or until non-exchangeable K+ was initiated to release. Thus, in the native soil, each leaching event consisted of the collection of 6650 ml leachate (95 Pv), equivalent to 3677 mm of rainfall or irrigation. For the Ca2+-saturated soil, each leaching event consisted of the collection of 3220 ml leachate (46 Pv), equivalent to 1783 mm of rainfall or irrigation. The pore volume of soil columns was calculated from the bulk density and particle density (2.65 g cm3) of soil in each column (e.g. Rowell, 1994) to be 70 ml. The solution level was maintained at approximately 5 cm above the soil surface within the soil column to maintain the effluent flow at an average of 1.570.1 ml min1. The water level varied slightly according to outflow rate. A filter paper was placed on the soil surface to minimize soil disturbance from the addition of leaching solution. Effluents from each leaching stage were collected in 35–70 ml lots; leachates were then analysed for soluble K+ and Ca 2+. The quantity of leached K+ was calculated from the K+ concentration and the volume of leachate fraction. The study was conducted with two replicates at room temperature (22–24 1C). 2.3. Isotherm experiments To determine K+ sorption, 2.5 g of soil was placed into a 50-ml centrifuge tube and mixed with 25 cm3 of varying equilibrating solutions with the following range of KCl concentrations: 0, 1, 3, 5, 10, 15, 20, 30, 40, and 50 mM K+. The equilibrating solutions contained 0, 3, 5, 10, and 15 mM CaCl2. Suspensions were shaken for 1 h. Sorbed K+ was calculated from the difference between the concentration of soluble K+ added to the initial solution and K+ in the solution at equilibrium (Rowell, 1994). This procedure was performed in duplicate. The high concentrations of dissolved K+ used in the sorption experiments reflect the high concentration of water-soluble K+ that occurs when K+ is placed in the top of the column. 3. Results and discussion 3.1. Leaching K+ from native soil The results of K+ leaching from native soil are presented as breakthrough curves (graphs showing the relationship between K+ concentration and cumulative water removed from the columns) in Fig. 1. Exchangeable K+ in the analysed soil, as extracted by 1 M neutral NH4OAc, was 0.52 cmolc kg1. The addition of CaCl2 caused the leaching of exchangeable K+ from the soil. Peak concentrations of K+ in the leachates were observed in approximately 1 Pv, followed by a sharp decrease in the subsequent leachate fractions

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

628

Concentration of K+ in the leachate (mg l-1)

45 15 mM CaCl2 40

10 mM CaCl2 5 mM CaCl2

35

3 mM CaCl2

30

Distilled water

25 20 15 10 5 0 0

20

40

60

80

100

Pore volumes Fig. 1. Breakthrough curves for potassium from the native soil leached with different CaCl2 solutions or distilled water.

with each treatment (Fig. 1). In soil leached with solutions containing low CaCl2 concentrations, only small amounts of K+ were released, with a steady decrease in K+ concentration as the experiments proceeded. Increased Ca2+ concentration in the leaching solution led to an increase in the concentration of K+ in the leachates. Table 1 shows the total amounts leached in up to 95 Pv. The amount leached varied from 18.4% to 69.7% of the exchangeable K+ when 20 Pv had passed through the column. The results show that soils irrigated with Ca-rich water can lose large amounts of K+, up to 371 kg ha1 for 95 Pv (in the 15 mM CaCl2 treatment). In the native soil leached with distilled water, 19 (5.7% exchangeable K+) and 61 (18.4% exchangeable K+) kg ha1 were leached with 5 and 20 Pv, respectively, because the supply of cations able to exchange with K+ was limited. The release of K+ was initially rapid, followed by a slower reaction. The first part of the experiment indicated rapid K+ release from the planar surface site, while the second part represented K+ release from edge and internal sites. The concentration of K+ in leachates after 75 Pv stabilized at 2.4 mg l1 (0.061 mM). The total leached amounts (Table 1) were in excess of initial exchangeable K+ (approximately 13% in 15 mM CalCl2), indicating the release of K+ from non-exchangeable sites. In kinetics studies of the same soil as that used in the present study using 10 mM CaCl2, the soil released 169 mg K+ kg1 over 2084 h (Jalali and Zarabi, 2006). The pattern of K+ leaching shown in Fig. 1 can be used to demonstrate the effects of poor-quality irrigation water in arid and semi-arid regions. Although Ca2+ solutions were used rather than the mixture of Na+, Ca2+, and Mg2+ normally present in natural waters (Rowell, 1994), the effect of water quality upon leaching is clear. 3.2. Displacement of a K+ pulse with CaCl2 from Ca2+-saturated soil To simulate fertilizer application in the field, a pulse of K+ was applied to the Ca2+-saturated soil in 36.9 ml of 50 mM KCl, equivalent to 400 kg K+ ha1. Increasing the

18.9

34.9

40.6

58.6

76.2

soil+distilled

soil+3 mM

soil+5 mM

soil+10 mM

soil +15 mm

23.2

17.8

12.4

10.6

5.7

144.5

110.5

76.2

66.2

33.3

44.0

33.6

23.2

20.1

10.1

% of ex. K

kg ha1

kg ha1

% of ex. Kb

Amount leached in 10 Pv (387 mm)

Amount leached in 5 Pva (193 mm)

229.0

192.8

141.6

125.3

60.6

kg ha1

69.7

58.7

43.1

38.1

18.4

% of ex. K

Amount leached in 20 Pv (774 mm)

Pv ¼ pore volume. The amount of exchangeable K+ in the native soil was equal to 328 kg ha1.

b

a

Native water Native CaCl2 Native CaCl2 Native CaCl2 Native CaCl2

Treatments

Table 1 Amounts of exchangeable K+ leached from columns of native soil

310.7

288.8

248.5

229.0



kg ha1

94.6

87.9

75.7

69.7



% of ex. K

Amount leached in 46 Pv (1780 mm)

346.7

329.8

291.5

271.8



kg ha1

105.6

100.4

88.8

82.8



% of ex. K

Amount leached in 71 Pv (2748 mm)

371.0

362.8

320.0

302.2



kg ha1

112.9

110.4

97.3

92.0



% of ex. Ka

Amount leached in 95 Pv (3677 mm)

Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

ARTICLE IN PRESS 629

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

630

CaCl2 concentration from 3 to 15 mM resulted in earlier breakthrough, a higher peak concentration of K+, and greater amounts of recovered K+ (Fig. 2). All breakthrough curves were of similar shape, with no indication of an extended leading edge that would result from by-pass flow, although the repacked 2 mm soil does not have the structural features that normally give rise to this effect (Wong et al., 1990). It is possible that the total leached amounts (Table 2) are not all derived from the applied K+, as native K+ had not been leached completely from the soil in the pre-treatment. Therefore, the actual recovery may be less than that shown in Table 2. For all CaCl2 solutions, the recovery of K+ was more than from the native soil (Table 1), indicating that K+ was leached more effectively when it was applied fresh to the surface rather than from exchange and mineral sources in the soil. The increased rate of K+ movement with greater Ca2+ concentration is related to the increased ability of Ca2+ to displace K+ from exchange sites into solution. Retardation of an adsorbed solute is known to depend on its adsorption coefficient (Wild, 1993); for K+ this coefficient decreases with increasing Ca2+ concentration. This effect will be further discussed later in the text. Changes in Ca2+ concentration during the transport of K+ are shown in Fig. 3. With the introduction of a pulse of K+, an exchange of Ca2+ occurs. The released Ca2+ moves freely with the applied chloride to appear after 1 Pv. Subsequently, the concentration of Ca2+ decreased, reaching its lowest value with the arrival of peak K+ before increasing to a stable concentration equal to that of the incoming CaCl2 solution. During leaching with CaCl2, the pulse with high initial K+ concentration was dispersed. Consequently, Ca2+ is readsorbed on exchange sites, leading to the low concentrations associated with the peak K+ value.

Concntration of K+ in the leachate (mg l-1)

180 160

15 mM CaCl2

140

10 mM CaCl2 5 mM CaCl2

120

3 mM CaCl2 Distilled water

100 80 60 40 20 0 0

10

20

30

40

50

Pore volumes Fig. 2. Breakthrough curves for a pulse of potassium in the Ca2+-saturated soils leached with different CaCl2 solutions or distilled water.

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

631

Table 2 Amounts of K+ leached from columns of Ca2+-saturated soil Treatments

Ca-saturated soil +distilled water Ca-saturated soil+3 mM CaCl2 Ca-saturated soil+5 mM CaCl2 Ca-saturated soil+10 mM CaCl2 Ca-saturated soil+15 mM CaCl2 a

Amount leached in 5 Pv (193 mm)

Amount leached in 10 Pv (387 mm)

Amount leaced in 20 Pv (774 mm)

Amount leached in 46 Pv (1780 mm)

kg ha1

kg ha1

kg ha1

kg ha1

% of applied Ka

% of applied K

% of applied K —



% of applied K

6.0

1.5

9.7

2.4





3.9

0.96

13.5

3.4

189.4

47.4

306.5

76.6

4.2

1.1

68.1

17.0

232.8

58.2

329.3

82.3

7.4

1.9

146.5

36.6

268.2

67.1

364.2

91.1

13.1

3.3

214.3

53.6

295.9

74.0

371.8

93.0

The amount of K+ applied to the column was equal to 400 kg ha1.

In all treatments, the two sides of the K+ breakthrough curves (Fig. 3) have different characteristics. The leading edge is sharp, while the following edge shows tailing. In soil leached with 15 mM CaCl2, the K+ concentration increased at a rate of 2.5 mg l1 per Pv for the first part of the breakthrough curve, while for the second part of the curve this rate was 1.27 mg l1 per Pv. It is possible that adsorption is rapid because of the high solution concentration in the applied pulse and that desorption is slow because of the smaller K+ gradient from exchange sites into solution. Other factors that influence the shape of these edges are solute reactions with clay minerals, organic matter, and sesquoxide (Cho et al., 1970; Selim et al., 1976; Cameron and Klute, 1977; Parker and van Genuchten, 1984). Sparks (1989) investigated the kinetics of K+ exchange on soils and clay minerals, and clearly demonstrated that the rate of exchange between solution and exchangeable K+ is diffusion controlled and strongly dependent on clay mineralogy. The soil analysed in the present study is mainly composed of illite, vermiculite, and smectites (Jalali, 2005a). The kinetics of K+ exchange on kaolinite and montmorillinite are usually quite rapid (Jardine and Sparks, 1984). For kaolinite, only planar surface sites are available for ionic exchange. For montmorillinite, interlayer sites are held together by hydrogen bonds, and layers are able to swell with adequate hydration; this enables the rapid passage of ions into the interlayer. The kinetics of K+ exchange on vermiculite and micas tend to be extremely slow. This slowness is attributed to internal sites on the surfaces of vermiculite and micas. High rates of reaction are commonly observed for external sites, intermediate rates for edge sites, and low rates for interlayer sites (Jardine and Sparks, 1984). From a physical viewpoint, sorption may be locally at equilibrium, but tailing can result from the diffusioncontrolled transfer of K+ between mobile and immobile liquid phases in the soil (van Genuchten and Wierenga, 1976; Nielsen et al., 1986).

ARTICLE IN PRESS

500

4

400

3

300

2

200

1

100

0

0 0

2

(a)

4

6

8

10

500

60 50

400

40

300

30

200

20 100

10

100

10

0 10

20

30

0 40

40

50

Pore volumes Ca2+

160

900

140

800

120

700 600

100

500

80

400

60

300

40

200

20

100 0

0

50

0

Pore volumes K+ Ca2+

(d)

180

1000

160

900

140

800 700

120

600

100

500 80

400

60

300

40

200

20

100

0

10

20

30

40

50

Pore volumes K+ Ca2+

Concentration of Ca2+ (mgl-1)

Concentration of K+ (mg l-1)

Concentration of K+ (mg l-1)

70

0

0 0

(e)

200

20

0

Concentration of Ca2+ (mgl-1)

Concentration of K+ (mg l-1)

600

(c)

300 30

0

700

30

400

40

K+

80

20

500

50

(b)

90

10

60

12

Pore volumes K+ Ca2+

0

600 Concentration of Ca2+ (mgl-1)

5

70

Concentration of Ca2+ (mg l-1)

600 Concentration of K+ (mgl-1)

Concentration of K+ (mg l-1)

6

Concentration of Ca2+ (mg l-1)

Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

632

10

20

30

40

50

Pore volumes K+ Ca2+

Fig. 3. Breakthrough curves for K+ and Ca2+ in Ca2+-saturated soils: (a) soil leached with distilled water, (b) soil leached with 3 mM CaCl2, (c) soil leached with 5 mM CaCl2, (d) soil leached with 10 mM CaCl2 and (e) soil leached with 15 mM CaCl2.

3.3. Displacement of a potassium pulse with distilled water in Ca2+-saturated soil Figs. 2 and 3 (a) shows the breakthrough curves for an experiment involving the displacement of a K+ pulse with distilled water in Ca2+-saturated soil. The peak concentration of Ca2+ recovered after 1.8 Pv and then decreased to a constant value of

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

633

about 33 mg l1 (0.8 mM) after 11 Pv. The soil contains CaCO3 (47 g kg1) that can solubilize to supply Ca2+ to replace K+ ions. Because of its extremely low solubility (0.0131 g l1) (Qadir et al., 1996), this native Ca2+ source does not make a significant contribution to the replacement of K+. In the field, the release of CO2 in the root zone that results from root and microbial respiration and CO2 dissolution in water produces H2CO3, which in turn increases CaCO3 dissolution. Hydrogen ions released in the root zone also lower the soil pH and can dissolve some CaCO3. Therefore, in the field, where CO2 concentration in the soil air is higher than that in the atmosphere, the dissolution of calcite can be enhanced. With more realistic field flow rates and CO2 concentrations, increased Ca2+ concentrations would lead to the increasingly rapid leaching of K+. After approximately 11 Pv, the flow rate became very slow and leaching ceased because of the loss of hydraulic conductivity. This experiment simulates the condition in the field where soil is irrigated with high-quality water or winter rain. A reduction in hydraulic conductivity occurs when the concentration of solution is reduced. The flow rate decreases as leaching progresses because of the removal of inherent salts by distilled water. This result is consistent with the expected response of colloid stability to a decrease in electrolyte concentration (Sposito, 1984). Generally, at an exchangeable sodium percentage (ESP) of between 10 and 15, soil clays are liable to swell and disperse. This leads to a deterioration in soil structure (Rowell, 1994), especially when the soil solution is diluted by rainwater or when high-quality (low salinity) irrigation water is applied (Shainberg and Letey, 1984). In the field, critical conditions occur when winter rain or high-quality irrigation water that enters the soil reduces the concentration of the soil solution to low values. Except at the soil surface, total ionic concentrations are maintained at about 3 mmolc l1, even during rain, as a result of buffering by exchangeable cations and the weathering of minerals such as feldespar, calcite, and gypsum (Rowell, 1994). At this concentration, the hydraulic conductivity of soils is affected even in soils with ESP values as low as 5. At the surface, low solution concentrations combined with the mechanical effects of raindrops can cause capping and reduce infiltration at even lower ESP values (Rowell, 1994). 3.4. Comparison of the leaching of K+ from native and Ca2+-saturated soil The amount of K+ leached by CaCl2 solution from native soil in which K+ was initially uniformly distributed was less than the amount removed from the Ca2+-saturated soil with the surface applied pulse. The native soil lost 41.4 mg of the 59.4 mg of K+ initially present in the column (69.7%) when leached with 20 Pv of 15 mM CaCl2 (Table 2). In the Ca2+saturated soil, 53.5 mg of the 72.3 mg initially applied (74%) was leached with the same volume of solution (Table 2). Only when leached with distilled water is the loss from the native soil greater than that from saturated soil (11 and 2.6%, respectively, with 11 Pv). In the native soil, K+ is distributed with a uniformly low concentration throughout the columns. Thus, a higher adsorption coefficient applies to this soil, resulting in a delay in leaching. When a pulse of K+ is applied to the top of the Ca2+-saturated soil column, a low adsorption coefficient results from the high concentration of K+; this enables more K+ to be leached. In addition, K+ within the native soil is not uniformly distributed over the exchange sites in a form that is readily available for exchange. Native K+ is much less readily released than recently adsorbed K+, as the former is more commonly trapped in the centre of interlayers and recently adsorbed K+ is adsorbed

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

634

on surface and wedge sites of the minerals. Potassium ions adsorbed on the planar surface are not specifically adsorbed, and the bonding is relatively weak, while K+ ions in the interlayers are very strongly bound, and K+ ions at the edges of the interlayer have intermediate bond strengths (Mengel, 1985). Potassium ions present in the interlayer and edge positions will only exchange readily with ions of the same size or smaller. As the hydrated radius of Ca2+ is 0.96 nm, it cannot readily replace K+ ions in interlayer sites; however, given sufficient time, replacement with Ca2+ results in an expansion of the clay mineral and the formation of wedge zones that contribute to the release of interlayer K+. 3.5. Isotherm experiments We calculated buffer powers (adsorption coefficients) from the isotherm experiments (Fig. 4). The slope of the adsorption isotherm provides the buffer power for K+ in the following units (Heming and Rowell, 1997): !, ! mg Kþ mg Kþ l , ¼ kg kg dry soil l in solution Measured values ranged from 1.016–0.473 l kg1 for 0–15 CaCl2 solutions (Table 3). As in the native soil, no K+ was added; accordingly, the lower part of the isotherm (0–5 mM) was used in the calculation. Equilibrium K+ concentrations of the soil (i.e. no net adsorption or sorption) ranged from 5–40.9 mgl1. Over this range, the plots were close to linear, and fixed values of buffer power were used. We found a linear correlation between equilibrium solution K+ concentration and leaching loss (Fig. 5). This relation can therefore be used for predicating leaching losses of K+ in soil leached with different CaCl2 concentrations. 2500

Adsorbed K+ (mg kg-1)

2000

1500

1000

500

0 0

500

1000

1500

2000

-500 equilibrium concnetration K+ (mg l-1) 0 mM CaCl2

3 mM CaCl2

5 mM CaCl2

10 mM CaCl2

15 mM CaCl2

Fig. 4. Potassium exchange isotherms for the Ca2+-saturated soil with a range of CaCl2 concentration.

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

635

Table 3 Adsorption coefficient, equilibrium K+ and proportion of exchangeable K+ in solution Treatments CaCl2 (mM)

r (g cm3)

y (cm3 cm3)

Potassium adsorption coefficienta (l kg1)

Potassium adsorption coefficientb (l kg1)

Equilibrium K+ (mg l1)

K+ in solution (%)

0 3 5 10 15

1.656 1.694 1.644 1.664 1.675

0.377 0.360 0.379 0.379 0.367

5.47 2.84 2.23 1.77 1.25

1.02 0.89 0.69 0.57 0.47

5.01 11.11 12.96 30.79 40.94

3.99 6.95 9.37 11.38 14.94

a

The slope of the adsorption isotherm over the range of 0–5 mM K+. The slope of the adsorption isotherm over the range of 0–50 mM K+.

b

The proportion of exchangeable K+ in soil solution is (Wild, 1981): 

1 , 1 þ br=y

b is the buffer power as measured for various CaCl2 concentrations (l kg1), r the bulk density of the soil (g cm3) and y the volumetric moisture content (cm3 cm3). The proportion of K+ in solution decreases with increased buffer power, increased bulk density, and decreased concentration of the leaching solution, while it increases with water content. Table 3 shows that 85–96% of the exchangeable K+ is adsorbed rather than remaining in solution. 3.6. Prediction of the delay in K+ leaching The simplest quantitative description of the delay of K+ leaching from a column of soil is the description used in chromatography, as with the following equation (Wild, 1981): Rf ¼ h



1  i , br y

where Rf is the retardation factor and other parameters are as defined earlier. As r and y are the same for all leaching columns, the prediction of delay requires knowledge of the adsorption coefficient (buffer power). This coefficient is dependent on the amount of negative charge and the concentrations of K+ and competing cations in solution (Wong et al., 1990; Jalali and Rowell, 2003). Predictions for these leaching experiments can be made using the term br/y, which is the ratio of adsorbed K+ to solution K+, both expressed per unit volume of soil (Wong et al., 1990). Any change that increases this ratio also acts to increase the delay. Thus, an increase in Ca2+ concentration will decrease K+ adsorption and therefore act to reduce the delay. There were delays of 7–13.5 Pv in the leaching of K+ with different CaCl2 concentrations (Fig. 2). The K+ delays calculated on the basis of adsorption coefficients obtained from the isotherms gave retardation factors of 0.34–0.19 for 0–15 mM CaCl2 treatments, respectively; this equates to delays of 2.9 and 5.3 Pv. In the columns, leaching was delayed more than predicted, indicating that the

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

636

y = 3.389x + 39.534 R2 = 0.9552

200 Leaching loss in 500 mm of simulated throughflow (kg K+ ha-1)

180 160 140 120 100 80 60 40 20 0 0

5

10

15

20

25

30 +

equilibrium concentration of K (mg

35

40

45

l-1)

Fig. 5. The relationship between equilibrium solution K+ and leaching loss after 500 mm of throughflow.

effective retardation factors were underestimated using this approach. Carrillo Gonzalez et al. (2004) used the same approach to predict the movement of Zn in columns and found that Zn was delayed more than predicted. Similar discrepancies were also reported by Wong et al. (1990) and Jalali (1997). 4. Conclusions Poor-quality ground-waters are a common feature of arid and semi-arid regions. When combined irrigation and rainfall exceeds the crop water requirement, excess soil water drains downward, carrying with it soluble salts including K+. The degree of K+ leaching is dependent on the Ca2+ concentration of the leaching waters: high-quality waters generate little displacement of K+. The application of CaCl2 in the reclamation of sodic soils would be expected to leach K+ from soils. In the studied area, Ca2+ is the dominant ion in well water, representing an average of 43.6% of total cations (Jalali, 2002a). The concentration of Ca2+ in irrigation water varies from 0.01 to 11.3 mM. Inputs of K+ into the irrigation water can balance the losses of leaching to some extent. The average concentrations of K+ in irrigation waters in the studied area are 5.5 mg l1 (Jalali, 2002a), while the average concentrations of K+ in irrigation water in the Bahar area (close to the studied area) is 2.5 mg l1 (Jalali, 2005b). Thus, 100 mm of irrigation water supplies 2.5–5.5 kg K+ ha1 compared to crop removals of, for example, 5 kg of K+ per tonne of wheat. Losses would generally be larger if the K+ is mixed throughout the ploughed layer rather than being applied as a pulse at the surface; however, with CaCl2 solutions, more K+ was leached from columns with a surface application than when the K+ was uniformly distributed. These results reflect the increased adsorption coefficients that apply for low K+ concentrations. In Ca2+-saturated soil, K+ concentrations in the leachate varied from 2 to 160 mg K+ l1, with a mean concentration of 23 mg K+ l1 for the different

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

637

treatments. Potassium concentrations in the leachate varied from 2.4 to 42.6 mg K+ l1, with a mean concentration of 12.1 mg K+ l1 for the various treatments in native soil. These concentrations are greater than the recommended guideline of the World Health Organization (12 mg K+ l1) (WHO, 1993). The column leaching experiment was not a perfect simulation of the field situation because the leaching rate was faster (49.773.3 mm h1) than that in the field; only about 4 h was required for 5 Pv to pass through the column. This was not sufficiently slow to ensure equilibration (the soil is sandy loam, and the same flow rate can be expected to occur in the field). When K+ fertilizer is applied in the field, more K+ is transferred into non-exchangeable forms. Experiments conducted on K+ fixation in sandy soils with similar clay content to the soil analysed in the present study demonstrate that 5–35% of the added K+ is transferred into non-exchangeable forms during 3 weeks of incubation (Jalali, 2002b); this would tend to reduce leaching losses. Additionally, the weathering of Kfeldspar, biotite, or muscovite provides a source of K+. In the field, the effects of stones must also be taken into account, as they act to intensify leaching, packing density, bypass flow, and input during times of rain (Heming and Rowell, 1997). The spreading of manure upon a calcareous soil may increase K+ desorption from exchange sites (Griffioen, 2001). The ammonification of manure followed by nitrification of the NH3 results in the production of nitrate and acid. Urea is also an important fertilizer in global agriculture and is widely used in arid and semi-arid soils. The hydrolysis of urea is rapid (Prakash et al., 1999). When urea is applied to a soil, it is hydrolysed to ammonium carbonate. The nitrification of NH+ 4 is also quite rapid. In carbonate-bearing soil, the acid produced by nitrification gives rise to an increase in the concentration of Ca2+ and Mg2+ in the soil solution. These ions then exchange with other cations, including K+, that were originally present in the soil. Therefore, an intensification of manure spreading and the application of urea to agricultural soils not only give rise to increased nitrate leaching but also increased associated K+ leaching. References Al-Wabel, M.A., Heil, D.M., Westfall, D.G., Barbarick, K.A., 2002. Solution chemistry influence on metal mobility in biosolids-amended soils. Journal of Environmental Quality 31, 1157–1165. Bar-Tal, A., Feigenbaum, S., Sparks, D.L., 1991. Potassium-salinity interactions in irrigated corn. Irrigation Science 12, 27–35. Beltran, J.M., 1999. Irrigation with saline water: benefits and environmental impact. Agricultural Water Management 40, 183–194. Cameron, D.R., Klute, A., 1977. Convective–dispersive solute transport with a combined equilibrium and kinetic and sorption model. Water Resources Research 13, 183–188. Carrillo Gonzalez, R., Rowell, D.L., Alloway, B.J., 2004. Displacement of Zn through acidic light-textured soils. Geoderma 124 (3–4), 335–348. Cho, C.M., Strong, J., Racz, G.J., 1970. Convective transport of orthophosphate (P-31 and P-32) in several Manitoba soils. Canadian Journal of Soil Science 50, 303–315. Feigenbaum, S., 1986. Potassium distribution in a sandy soil exposed to leaching with saline water. In: Nutrient Balances and the Need for Potassium. International Potash Institute, Reims, pp. 155–162. Feigenbaum, S., Meiri, A., 1988. The effect of potassium fertilization on cotton response and potassium distribution under irrigation with saline water. BARD Report I-630-83, 88–110. Griffioen, J., 2001. Potassium adsorption ratios as an indicator for the fate of agricultural potassium in groundwater. Journal of Hydrology 254, 244–254. Heming, S.D., Rowell, D.L., 1997. The estimation of losses of potassium and magnesium from chalky soils: laboratory studies. Soil Use and Management 13, 122–129.

ARTICLE IN PRESS 638

Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

Jalali, M., 1997. Measuring and modelling the leaching of potassium in a sandy soil. Ph.D. Thesis. Reading University, Reading, UK. Jalali, M., 2002a. Composition of irrigation waters in west of Iran. In: The 17th World Congress of Soil Science in Bangkok, Thailand, 14–21 August, pp. 2184-1–2184-4. Jalali, M., 2002b. The effect of potassium fertilizer on the soil solution and exchangeable potassium. Iranian Journal of Soil and Water Sciences 16 (1), 1–12. Jalali, M., 2005a. Release kinetics of non-exchangeable potassium in calcareous soils. Communications in Soil Science and Plant Analysis 36, 1903–1917. Jalali, M., 2005b. Major ion chemistry in the Bahar area, Hamadan, Western Iran. Environmental Geology 47, 763–772. Jalali, M., Rowell, D.L., 2003. The role of calcite and gypsum in the leaching of potassium in a sandy soil. Experimental Agriculture 39, 379–394. Jalali, M., Zarabi, M., 2006. Kinetics of nonexchangeable-potassium release and plant response in some calcareous soils. Journal of Plant Nutrition and Soil Science 169 (2), 196–204. Jardine, P.M., Sparks, D.L., 1984. Potassium–calcium exchange in a multireactive soil system. I Kinetics. Soil Science Society of America Journal, 48, 39–45. Johnston, A.E., Goulding, K.W.T., 1992. Potassium concentrations in surface and groundwater and the loss of potassium in relation to land use. In: Potassium in Ecosystems, Biogeochemical Fluxes of Cations in Agro-and Forest-Systems. International Potash Institute, Basel, pp. 35–158. Johnston, A.E., Goulding, K.W.T., Mercer, E., 1993. Potassium leaching from a sandy soil. International Potash Institute, Subject 12, No. 4. Martin, H.W., Sparks, D.L., 1985. On the behaviour of nonexchangeable potassium in soils. Communications in Soil Science and Plant Analysis 16, 133–162. Marwan, M.M., Rowell, D.L., 1995. Cation exchange, hydrolysis and clay movement during displacement of saline solutions from soils by water. Irrigation Science 16, 81–87. Meiri, A., Feigenbaum, S., Sagiv, B., 1984. Potassium fertilization under irrigation with saline and sodic water. Report to Dead Sea Works, 301-00-81. Mengel, K., 1985. Dynamics and availability of major nutrients in soils. Advances in Soil Science 2, 65–131. Minhas, P.S., 1996. Saline water management for irrigation in India (review). Agricultural Water Management 30, 1–24. Nielsen, D.R., van Genuchten, M.Th., Biggar, J.W., 1986. Water flow and solute transport processes in the unsaturated zone. Water Resources Research 22, 89S–108S. Parker, J.C., van Genuchten, M.Th., 1984. Determining transport parameters from laboratory and field tracer experiments. Bulletin 84-3, Virginia Agriculture Experiment Station, Blackburg. Prakash, O., Alva, A.K., paramasivam, S., 1999. Use of the urease inhibitor N-(n-butyl)-thiophosphoric triamide decreased nitrogen leaching from urea in a fine sandy soil. Water, Air, and Soil Pollution 116, 587–595. Qadir, M., Qureshi, R.H., Ahmad, N., 1996. Reclamation of a saline-sodic soil by gypsum and Leptochloa fusca. Geoderma 74, 207–217. Rhoades, J.D., Kandiah, A., Mishali, A.M., 1992. The use of saline waters for crop production. FAO Irrigation and Drainage Paper No. 48. FAO, Rome, 133pp. Rowell, D.L., 1985. The reduction in sodicity during the displacement of mixed CaCl2–NaCl salts from soils by water. Irrigation Water 6, 11–18. Rowell, D.L., 1994. Soil Science: Methods and Applications. Longman Group, Harlow. Rowell, D.L., Dimitrios, P., 2002. Diffusion and cation exchange during the reclamation of saline-structured soils. Geoderma 107, 271–279. Selim, H.M., Mansell, R.S., Zelanzny, L.W., 1976. Analytical solution of the equation of reactive solutes through soils. Water Resources Research 12, 528–532. Shainberg, I., Rhoades, J.D., Prather, R.J., 1981. Effect of low electrolyte concentration on clay dispersion and hydraulic conductivity of a sodic soil. Soil Science Society of America Journal 45, 273–277. Shainberg, I., Letey, J., 1984. Response of soils to sodic and saline conditions. Hilgardia 52, 1–57. Sparks, D.L., Huang, P.M., 1985. Physical chemistry of soil potassium. In: Munson, R.D. (Ed.), Potassium in Agriculture. American Society of Agronomy, Madison, WI, pp. 202–264. Sparks, D.L., 1989. Soil Chemistry: Kinetics and Mechanisms. McGraw-Hill Yearbook Science and Engineering, McGraw-Hill, New York, pp. 362–365. Sposito, G., 1984. The Surface Chemistry of Soils. Oxford University Press, New York.

ARTICLE IN PRESS Z. Kolahchi, M. Jalali / Journal of Arid Environments 68 (2007) 624–639

639

Tanji, K.K. (Ed.), 1990. Agricultural salinity—assessment and management. ASCE Manual No. 71. ASCE, New York, 619pp. van Genuchten, M.Th., Wierenga, P.J., 1976. Mass transfer studies in sorbing porous media. I. Analytical solutions. Soil Science Society of America Journal 40, 473–480. WHO, 1993. Guidelines for drinking water quality. 1. Recommendations, second ed. World Health Organisation, Geneva. Wild, A., 1981. Mass flow and diffusion. In: Greenland, D.J., Hayes, M.H.B. (Eds.), The Chemistry of Soil Processes. Wiley Interscience, Chichester, pp. 37–80. Wild, A., 1993. Soils and The Environment: An Introduction. Cambridge University Press, Cambridge. Wong, M.T.F., Hughes, R., Rowell, D.L., 1990. Retarded leaching of nitrate in acid soils from the tropics: measurement of the effective anion exchange capacity. Journal of Soil Science 41, 655–663.