Potassium and magnesium in irrigation water quality

G Model

ARTICLE IN PRESS

AGWAT-3993; No. of Pages 6

Agricultural Water Management xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Potassium and magnesium in irrigation water quality assessment C.J. Smith a,∗ , J.D. Oster b , G. Sposito c a

CSIRO Land and Water, PO Box 1666, Canberra ACT 2601, Australia Department of Environmental Science, University of California, Riverside, CA 92506, USA c Department of Environmental Science, Policy and Management, University of California, Berkeley, CA 94720-3114, USA b

a r t i c l e

i n f o

Article history: Available online xxx Keywords: Waste water Hydraulic conductivity Cations Stability index

a b s t r a c t There is a growing trend of K replacing Na in wastewaters from agricultural industries to reduce the adverse effects of irrigation with these waters on soil hydraulic properties. However, problems with soil physical properties caused by wastewaters with high concentrations of K have been reported in both Australia and California. A review of the literature dating back to the 1930s supports the general conclusion that the relative order of deleterious effect on soil hydraulic properties of the four common cations in soils is Na > K > Mg > Ca. This paper explores the capabilities of the Cation Ratio of Structural Stability (CROSS), a generalized Sodium Adsorption Ratio incorporating all four cations, as an improved irrigation water quality parameter. This new parameter includes a coefficient for K that accounts for its relative deleterious effect as compared with Na and another coefficient for Mg that accounts for its relative beneficial effect as compared with Ca. Based on optimizing CROSS using threshold electrolyte concentration data for a Sodosol from the Riverina Region of Australia, the deleterious effect of K is estimated to be about one-third of that of Na, while the concentration of Mg needs to be about an order of magnitude larger than Ca to have the same beneficial effect. © 2014 Published by Elsevier B.V.

1. Introduction Wastewaters generated by agriculture and municipalities are being re-used increasingly for irrigation, both as a way to mitigate severe water shortages and as means of avoiding wastewater discharge to surface waters while producing economic value from wastewater streams (Bond, 1998; Arienzo et al., 2012). However, the sustainable use of wastewaters generated by agriculture poses significant challenges because they can contain high concentrations of K. This is especially the case for wastewaters generated by dairies, piggeries, wineries, and olive or tomato processing plants (Smiles and Smith, 2004a; Arienzo et al., 2009, 2012; Dellavalle, 2010, personal communication; Rengasamy and Marchuk, 2011). The change in water composition arises mainly as a result of moving from Nato K-based products in the cleaning process and/or shifting animal diet, wherein cheese whey (K concentration: 1660 mg/L; Feigin et al., 1991) is used as part of the feed formulation (intensive piggeries). In many cases, the shift to K has been deliberate because of well-established concerns about negative effects of Na on soil physical properties (U.S. Salinity Laboratory Staff, 1954; Levy, 2012).

∗ Corresponding author. Tel.: +61 262465960. E-mail address: [email protected] (C.J. Smith).

However, recent studies have documented problems with water infiltration caused by high levels of K in applied waters (Rengasamy and Marchuk, 2011), particularly in Australia (Smiles and Smith, 2004a) and California (Nat Dellavalle, personal communication, 2010). Generally, studies of the hydraulic properties of salt-affected soils have focused exclusively on the potential deleterious effects of two soil properties, exchangeable Na and soil water salinity (Quirk and Schofield, 1955; Bresler et al., 1982; Shainberg and Shalhevet, 1984; Sumner, 1993; Sumner and Naidu, 1998; Oster and Jayawardane, 1998; Levy, 2012). One reason the effect of K on soil hydraulic properties has not traditionally been taken into consideration is that Na concentrations in salt-affected soils and groundwaters are usually much higher than those of K, but the most important reason is that the authors of Handbook 60 (U.S. Salinity Laboratory Staff, 1954) concluded 60 years ago that “exchangeable K has only a slight or no adverse effect upon the physical properties of soils.” This conclusion was influenced by “measurements made recently at the Laboratory on samples of seven soils adjusted to various levels of exchangeable sodium and exchangeable potassium (Fig. 1)”. The figure in question displays the ratio of air permeability to water permeability as a function of exchangeable sodium percentage (ESP) and exchangeable potassium percentage (EPP). This ratio increases exponentially with ESP, whereas for EPP there

http://dx.doi.org/10.1016/j.agwat.2014.09.003 0378-3774/© 2014 Published by Elsevier B.V.

Please cite this article in press as: Smith, C.J., et al., Potassium and magnesium in irrigation water quality assessment. Agric. Water Manage. (2014), http://dx.doi.org/10.1016/j.agwat.2014.09.003

G Model AGWAT-3993; No. of Pages 6 2

ARTICLE IN PRESS C.J. Smith et al. / Agricultural Water Management xxx (2014) xxx–xxx

is no increase observed for three of the seven soils examined, while the increase is small for the other four. Variations of this figure were published by Reeve et al. (1954) and Brooks et al. (1956), who stated: “Combinations of various levels of exchangeable sodium and potassium showed that exchangeable potassium, whether alone or in the presence of exchangeable sodium, had little if any effect in increasing the permeability ratio. . .” One may thus infer that the authors of Handbook 60 were confident in setting aside a premise stated long ago by de Sigmond (1928) and Magistad (1945), that exchangeable Na and K should be considered to have additive deleterious effects on soils. Even today, the potential impact of exchangeable K on soil properties is not mentioned in ASCE No. 71, a standard engineering manual related to salinity assessment and management (Wallender and Tanji, 2012).

2. Effects of K and Mg on soil hydraulic conductivity Quirk and Schofield (1955), inspired by research on the effects of electrolyte concentration on the permeability of soils in California (Fireman and Bodman, 1939), reported what appears to be the first systematic investigation to document the effects of K on the saturated soil hydraulic conductivity. They equilibrated soil pads with concentrated Cl solutions of Na, K, Ca, or Mg, then leached them with a series of more dilute Cl solutions of the same cation. Their results showed decreases in the hydraulic conductivity over a fivehour period of leaching which depended on both the cation and the leaching solution concentration. The decrease in hydraulic conductivity followed the order: Na > K > Mg > Ca. Quirk and Schofield (1955) defined the cation concentration at which a 10–15% reduction in the saturated hydraulic conductivity occurred at the end of the leaching period as the threshold (electrolyte) concentration (TEC), reporting TEC values for Na, K, Mg, and Ca as 250, 67, 2.0 and 0.6 mmolc /L, respectively. They noted in particular for K that a concentration of 20 mmolc /L produced turbidity in the soil percolate, and a concentration of 10 mmolc /L resulted in clay deflocculation. Since this early work, researchers have reported both positive and negative effects of K on soil hydraulic conductivity. Keren (1984), in his comprehensive review of the effects of K and Mg on soils, concluded that these mixed results indicate “the effect of exchangeable K on soil properties is not well understood.” The same ambivalence was echoed recently in a brief review by Rengasamy and Marchuk (2011), who also noted that the need to solve this conundrum has grown more urgent because of increasing irrigation with wastewaters. Levy and van der Watt (1990), working with three South African soils, found little K effect on an Alfisol with kaolinite and iron oxide mineralogy (EPP of 6 or 13), a much larger effect on an Aridisol with illite and smectite mineralogy (EPP of 18.9), and an intermediate effect on an Alfisol with this same mineralogy. For two soils having an illite content of 10–11%, Chen et al. (1983) reported that the saturated hydraulic conductivity increased with increasing EPP < 20%, while for EPP > 20%, the conductivity decreased with increasing EPP. However, for a third soil which had an illite content of 16%, the hydraulic conductivity uniformly decreased with increasing EPP. Scanning electron micrographs showed that enrichment with K resulted in the formation of dense, continuous networks of clay particles filling the pore spaces between sand particles (Chen et al., 1983). Marchuk et al. (2012) applied computed X-ray tomography to characterize changes in pore architecture as influenced by exchangeable Na, K, Mg, and Ca. Total porosity and pore connectivity were influenced by cation valence, as might be anticipated from the Schulze-Hardy Rule (Sposito, 2008). A well-developed

pore structure was evident in soils having a sufficiently high concentration of exchangeable Ca and Mg, whereas soils dominated by Na and K had isolated pore clusters surrounded by pores filled with dispersed clay particles (Marchuk et al., 2012). Clay dispersion decreased in the order: Na > K > Mg > Ca and, correspondingly, hydraulic conductivity decreased in the reverse order, as reported originally by Quirk and Schofield (1955). Magnesium has long been considered to have effects similar to Ca on soil physical properties (Keren, 1984; Levy, 2012), leading the U.S. Salinity Laboratory Staff (1954) to group these two bivalent cations together as promoting and maintaining good soil structure. Bresler et al. (1982) remarked that this customary grouping may not reflect the true status of Mg, which is masked by the typically twoto five-fold greater concentration of Ca over Mg in irrigation waters and soil solutions. Rengasamy et al. (1984) concluded that Mg has the greatest effect on the saturated hydraulic conductivity when the electrolyte concentration is low, a point also made by Keren (1984), Sumner (1993), and Levy (2012). Use of gypsum on silt loam soils with an exchangeable Mg saturation near 36% reduced runoff under surface irrigation conditions (Vyshpolsky et al., 2010). In experiments conducted on soils from the corn belt in the United States, Dontsova and Norton (2002) reported Mg-treated soils had half the final infiltration rate as Ca-treated soils for two of four soils they studied. In summary, the available literature supports the premise that, under certain conditions, both K and Mg can have deleterious effects on soil hydraulic properties, and when there are deleterious effects, they usually relate to clay particle dispersion, resulting in reduced soil permeability. These trends, observed for soils of varying clay mineralogy, have to do with the fundamental relationship between soil particle flocculation and cation adsorption. If we think of the simplest theoretical approach to this relationship, diffuse double layer theory comes to mind (Sposito, 2008). This theory implies that only cation valence (Z) matters, and hence all cations of a given valence should behave in the same way, although monovalent cations should behave differently from bivalent cations. Therefore, one parameter to explain particle flocculation is cation valence, and this parameter alone suffices not only to understand the Schulze-Hardy Rule, but also to develop a semi-quantitative model of critical flocculation concentrations (Sposito, 2008). But such a model cannot explain Na–Mg systems as compared to Na–Ca systems, or K–Ca versus Na–Ca. Two other parameters that have been used to account for cation adsorption phenomena are ionic potential (Z/R) and Misono softness (Sposito, 2008). The first one (ionic potential, Z/R) scales the valence by cation radius (R) and is a measure of the electrostatic binding energy of a cation, whereas the second one (Misono softness) involves the energy required to extract an electron from a cation and is a measure of the covalency of cation binding. Recently Marchuk and Rengasamy (2011) combined these two parameters into a single index for the degree of covalency of cation binding, then subtracted it from 1.0 to get what they termed the degree of non-covalency, or the ionicity index. This new index depends on cation valence and the energy required to extract an electron from a cation, but it does not depend on cation size. Thus it goes beyond diffuse double layer theory by saying that the degree of covalency in cation binding also is important, not just electrostatics, and in this way it distinguishes Na from K, and Ca from Mg. Table 3 in Marchuk and Rengasamy (2011) shows that the ionicity index is a very good predictor of both clay suspension turbidity and clay particle zeta potential for a wide variety of soil clays. In particular, it predicts observed differences in turbidity, a measure of clay dispersion, with the ordering: Na > K > Mg > Ca. We suggest, therefore, that the reason for the observed differences in soil response to applied water composition, summarized above, is related to both the valence of the cations and their ionicity index.


G Model AGWAT-3993; No. of Pages 6


3

3. A new water quality parameter related to soil hydraulic properties

is 0.56 (1/1.8). The coefficient of Mg, a measure of its flocculating power relative to Ca, is 0.60 (27/45), resulting in the definition:

3.1. Moving beyond SAR

CROSSf =

The Sodium Adsorption Ratio (SAR) has long been the standard measure of potential sodium hazard for irrigation water (U.S. Salinity Laboratory Staff, 1954):

where we have added a subscript f to indicate that the coefficients are based on the relative flocculating power. Rengasamy and Marchuk (2011) tested Eq. (3) by correlating SAR, MCAR, and CROSSf with their measurements of the percent dispersible clay in four Australian soils they examined. Although all three water quality parameters were correlated significantly with percent dispersible clay, the correlation with CROSSf was greatly superior to that with SAR and somewhat better than that with MCAR. They also found that CROSSf showed better correlation than MCAR with the ratio of exchangeable monovalent to exchangeable bivalent cations in the four Australian soils. In light of the fact that the electrolyte concentration required to cause flocculation of soil clays is usually considerably greater than that which results in the dispersion of soil clays (Quirk, 2001), we used the TEC values reported by Quirk and Schofield (1955), which relate to clay dispersion, instead of CFC values to calculate the two coefficients in CROSS. Thus, comparing clay dispersion instead of flocculation for the monovalent cations, the coefficient of K would be the ratio of TEC for K to that for Na, 0.26 (66/250). For the bivalent cations we must compare flocculation, so the coefficient of Mg would be the ratio of TEC for Ca to that for Mg, 0.3 (0.6/2.0), leading to the final expression:

SAR =

Na [(Ca + Mg)/2]

0.5

(1)

where a chemical element symbol indicates a concentration in mmolc /L. The SAR can be related through a rigorous thermodynamic argument to the exchangeable sodium percentage (Oster and Sposito, 1980), a key soil property determining aggregate stability and soil permeability (Bresler et al., 1982; Keren, 1984; Levy, 2012). Similarly, the potassium adsorption ratio (PAR) has been defined, with K concentration replacing that of Na (U.S. Salinity Laboratory Staff, 1954), but there are no guidelines based on PAR in the standard reference publications related to irrigation water quality assessment (Ayers and Westcot, 1985; Rhoades et al., 1992; Tanji and Kielen, 2002; Tyagi and Minhas, 1998; Wallender and Tanji, 2012). However, recent research (Smiles and Smith, 2004a,b; Kruger et al., 1995; Arienzo et al., 2009; Rengasamy and Marchuk, 2011) has reported a reduction in saturated hydraulic conductivity from the accumulation of exchangeable K. Smiles and Smith (2004a), recognizing the possibility of a synergistic effect of monovalent cations on soil hydraulic properties, proposed that the SAR be replaced by the Monovalent Cation Adsorption Ratio (MCAR) to assess irrigation water quality:

MCAR =

Na + K + NH4 [(Ca + Mg)/2]

0.5

(2)

These authors pointed out the need to apply MCAR carefully, because K can become strongly adsorbed in soils, whereas NH4 may be only ephemeral in soil, becoming rapidly oxidized to NO3 . Indeed, under conditions favourable for rapid nitrification, Intrawech et al. (1982) found that ammonium-based fertilizers had no measurable influence on soil structure or compactability during a 10-year field study conducted in Kansas under cropped conditions. In the MCAR, no relative distinction is made between Na and K in promoting soil clay dispersion, or between Ca and Mg in promoting soil clay flocculation. Considering differences in the flocculating power of cations having the same valence (Sposito, 2008), Rengasamy and Marchuk (2011) proposed an alternative to the MCAR that incorporates the differing effects of Na and K as dispersing cations and of Mg and Ca as flocculating cations. This new irrigation water quality parameter, the Cation Ratio of Structural Stability (CROSS), incorporates the inverse of the critical flocculation concentration (CFC) for a particular cation as a measure of its flocculating power. Rengasamy and Sumner (1998) have reported CFC values for K-, Mg-, and Ca-saturated clays extracted from four soils, which they used to calculate the average relative flocculating power of each cation by dividing its average CFC into the average CFC for Na-clay, taken as a reference. Compared to Na, the average relative flocculating power of K, Mg, and Ca for the four soils was found to be: 1.8 ± 0.3, 27 ± 5, and 45 ± 8, respectively. CROSS differs from MCAR by introducing ratios of the relative flocculating power as coefficients of the K and Mg concentrations in Eq. (2), after dropping NH4 (Rengasamy and Marchuk, 2011). The coefficient of K, a measure of its dispersing power relative to Na,

Na + 0.56K [(Ca + 0.60Mg)/2]

CROSSd =

0.5

Na + 0.26K [(Ca + 0.30Mg)/2]

0.5

(3)

(4)

where the subscript d indicates that the coefficients for K and Mg are now based on relative dispersing power. We tested Eq. (4) using laboratory data reported by Jayawardane et al. (2011) and Arienzo et al. (2012) for a Sodosol from the Riverina region of Australia which had been irrigated with winery wastewater. This soil is high (>50%) in smectite clay minerals and has alkaline pH (>8), with clay loam texture overlying a medium clay texture in its subsurface layer. The TEC values we calculated with these data, as defined according to Quirk and Schofield (1955), are listed in Table 1. Linear correlations between MCAR, CROSSf , or CROSSd and these TEC values are shown in Fig. 1. The correlation between MCAR and TEC (Fig. 1A) was very poor. Considerable improvement occurred when the correlation was between CROSSf and TEC (Fig. 1B), and that between CROSSd and TEC was even better (Fig. 1C). These results confirm that K effects on TEC as a dispersing cation are not the same as those of Na, and Mg effects on TEC as a flocculating cation are not the same as those of Ca. 3.2. Optimizing CROSS At this juncture, it is useful to step back and reflect on the significance of the two coefficients appearing in CROSS. This consideration is facilitated by generalizing the definition of CROSS to include both Eqs. (3) and (4) as special cases: CROSS

=

Na + aK

[(Ca + bMg)/2] = SAR∗ + aPAR∗

0.5

(5)

where a and b are now coefficients to be determined, while SAR∗ =

Na [(Ca + bMg)/2]

0.5

(6)

and PAR∗ =

K [(Ca + bMg)/2]

0.5

(7)



ARTICLE IN PRESS C.J. Smith et al. / Agricultural Water Management xxx (2014) xxx–xxx Table 1 Sodium (SAR) and potassium adsorption ratio (PAR), threshold concentration (TEC, mmolc /L) and cation concentrations (mmolc /L) in applied water leading to a 15% reduction in the saturated hydraulic conductivity of surface and subsurface layers in a calcareous soil from the Riverina region of Australia (Arienzo et al., 2012). Soil layer

SAR or PAR

TEC

K

Na

Mg

Ca

Surface Surface Surface Surface Surface Surface Subsurface Subsurface Subsurface Subsurface Subsurface Subsurface Surface Surface Surface Surface Surface Surface Subsurface Subsurface Subsurface Subsurface Subsurface Subsurface Surface Surface Surface Surface Surface Surface Subsurface Subsurface Subsurface Subsurface Subsurface Subsurface

SAR40 Ca PAR40 Ca SAR20 Ca PAR20 Ca SAR5 Ca PAR5 Ca SAR40 Ca PAR40 Ca SAR20 Ca PAR20 Ca SAR5 Ca PAR5 Ca SAR40 Ca+Mg PAR40 Ca+Mg SAR20 Ca+Mg PAR20 Ca+Mg SAR5 Ca+Mg PAR5 Ca+Mg SAR40 Ca+Mg PAR40 Ca+Mg SAR20 Ca+Mg PAR20 Ca+Mg SAR5 Ca+Mg PAR5 Ca+Mg SAR40 Mg PAR40 Mg SAR20 Mg PAR20 Mg SAR5 Mg PAR5 Mg SAR40 Mg PAR40 Mg SAR20 Mg PAR20 Mg SAR5 Mg PAR5 Mg

66.0 15.5 30.2 9.6 2.0 2.0 134.0 32.9 16.7 26.9 1.5 1.5 191 22.6 66.9 10.7 1.5 2 224.8 71.9 33.3 7.1 1 1 517.0 140.0 283.0 99.5 23.2 18.9 501.0 136.0 156.0 60.7 8.7 15.1

0.00 15.21 0.00 9.18 0.00 1.76 0.00 31.65 0.00 24.92 0.00 1.35 0.00 22.00 0.00 10.18 0.00 1.76 0.00 66.40 0.00 6.87 0.00 0.93 0.00 121.50 0.00 73.00 0.00 10.30 0.00 118.50 0.00 49.00 0.00 8.85

61.30 0.00 26.70 0.00 1.76 0.00 116.90 0.00 15.50 0.00 1.35 0.00 159.30 0.00 52.95 0.00 1.35 0.00 183.00 0.00 29.10 0.00 0.93 0.00 357.00 0.00 158.00 0.00 12.00 0.00 349.00 0.00 103.00 0.00 5.90 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 15.85 0.30 6.98 0.26 0.07 0.12 20.90 2.75 2.10 0.12 0.03 0.03 160.00 18.50 125.00 26.50 11.20 8.60 152.00 17.50 53.00 11.70 2.80 6.25

4.70 0.29 3.50 0.42 0.24 0.24 17.10 1.25 1.20 1.98 0.15 0.15 15.85 0.30 6.98 0.26 0.07 0.12 20.90 2.75 2.10 0.12 0.03 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

are generalizations of SAR and PAR, respectively. Eq. (5) indicates that CROSS can be interpreted as a weighted sum of a generalized SAR and PAR, with the weighting factor a serving as a measure of the efficacy of PAR*, relative to that of SAR*, as a control on the soil property with which they are correlated. In particular, the linearity of Eq. (5) implies that the y-intercept of a linear correlation between CROSS and TEC, i.e., the limiting value of CROSS as TEC becomes very small, will scale with the value of the coefficient a. If a < 1, as it is in Eqs. (3) and (4), then the impact of PAR* on increasing the TEC is accordingly less than that of SAR*. On the other hand, the slope of a correlation between CROSS and TEC should scale with the inverse-square-root of the coefficient b, implying that the coefficient b cannot be interpreted in quite the same way as a. By analogy with activity coefficients in chemical thermodynamics, one could interpret the product bMg as an effective concentration of Mg, analogous to the product of an activity coefficient and a molar concentration of an ion defining its effective concentration. In the present context, the effective concentration refers to the relative impact of Mg in promoting soil clay flocculation or, equivalently, in preventing clay dispersion, as compared to Ca taken as a reference. Thus b is expected to be K > Mg > Ca. However, these important differences between the two monovalent and the two divalent ions are not currently taken into account in existing irrigation water quality criteria, which are based on salinity and the SAR of the irrigation water. The concept of CROSS, which has been proposed as an index of irrigation water quality to replace SAR, takes these differences into account. The coefficients for K and Mg in the original formulation of CROSS (Marchuk and Rengasamy, 2010) were based on the clay flocculating ability of Mg as compared to Ca, or K as compared to Na. Whilst this concept is reasonable, we found that coefficients based on clay dispersing ability, i.e., TEC values for the four cations at which the saturated hydraulic conductivity begins to decline significantly, improved the correlation between CROSS and TEC for a Sodosol soil from the Riverina region of Australia irrigated with winery wastewater. This result led us to postulate a more general definition of CROSS (Eq. (5)), which produced the best correlation with TEC for this soil through optimization of the

We consider our individual contributions to this paper to be equal. The authors thank Dr. Jayawardane for providing the raw data on relative hydraulic conductivity changes published in Figs. 1 and 2 in the papers by Jayawardane et al. (2011) and Arienzo et al. (2012). Gratitude also is expressed to Dr. A. L. Page for helpful comments on a draft version of this paper. Support for Dr. Smith was provided by the Australian Water Recycling Centre of Excellence under the Commonwealth’s Water for the Future Program and the CSIRO Water for a Healthy Country Research Flagship. Preparation of this paper also was supported in part by funds allocated to Dr. Sposito as the Betty & Isaac Barshad Chair in Soil Science and administered through the College of Natural Resources, University of California at Berkeley.

References Arienzo, M., Christen, E.W., Quayle, W.C., Kumar, A., 2009. A review of the fate of potassium in the soil–plant system after land application of wastewaters. J. Hazard. Mater. 164, 415–422. Arienzo, M., Christen, E.W., Jayawardane, N.S., Quayle, W.C., 2012. The relative effects of sodium and potassium on soil hydraulic conductivity and implications for winery wastewater management. Geoderma 173–174, 303–310. Ayers, R.S., Westcot, D.W., 1985. Water quality for agriculture. FAO Irrig. Drain. Pap. 29 Rev.1. FAO, Rome. Bond, W.J., 1998. Effluent irrigation – an environmental challenge for soil structure. Aust. J. Soil Res. 36, 543–555. Bresler, E., McNeal, B.L., Carter, D.L., 1982. Saline and Sodic Soils. Springer-Verlag, Berlin, Germany. Brooks, R.H., Bower, C.A., Reeve, R.C., 1956. The effect of various exchangeable cations upon the physical condition of soils. Soil Sci. Soc. Am. J. 20, 325–327. Chen, Y., Banin, A., Borochovitch, A., 1983. Effects of potassium on soil structure in relation to hydraulic conductivity. Geoderma 30, 135–147. de Sigmond, A.A.J., 1928. The classification of alkali and salty soils. 1st Int. Cong. Soil Sci. Proc. (1927) 1, 330–344. Dellavalle, N., 2010. Private Communication. Dellavalle Laboratory Inc., 1910, W McKinley Ave, Fresno, CA, pp. 93728. Dontsova, K.M., Norton, L.D., 2002. Clay dispersion, infiltration, and erosion as influenced by exchangeable Ca and Mg. Soil Sci. 167, 184–193. Duan, Q.Y., Gupta, V.K., Sorooshian, S., 1993. Shuffled complex evolution approach for effective and efficient global minimization. J. Optim. Theory Appl. 76 (3), 501–521. Feigin, A., Ravina, I., Shalhevet, J., 1991. Irrigation With Sewage Effluents. Advanced Series in Agricultural Science, vol. 17. Springer, Berlin. Fireman, M., Bodman, G.B., 1939. The effect of saline irrigation water upon the permeability and base status of soils. Soil Sci. Soc. Am Proc. 4, 71–77. Intrawech, A., Stone, L.R., Ellis Jr., R., Whitney, D.A., 1982. Influence of fertilizer nitrogen source on soil physical and chemical properties. Soil Sci. Soc. Am. J. 46, 832–836. Jayawardane, N.S., Christen, E.W., Arienzo, M., Quayle, W.C., 2011. Evaluation of the effects of cation combinations on soil hydraulic conductivity. Soil Res. 49, 56–64. Keren, R., 1984. Potassium, magnesium and boron in soils under saline and sodic conditions. In: Shainberg, L., Shalhevet, J. (Eds.), Soil Salinity under Irrigation: Processes and Management. Springer-Verlag, Berlin, Germany, pp. 77–99. Kruger, I., Taylor, G., Ferrier, M., 1995. Effluent at work. Australian Pig Housing Series, NSW Agriculture, Tamworth, NSW. Levy, G.J., 2012. Sodicity. In: Huang, P.M., Li, Y., Sumner, M.E. (Eds.), Handbook of Soil Sciences. Resource Management and Environmental Impact, 2nd ed. CRC Press, Boca Ratón, FL (Chapter 18). Levy, G.J., van der Watt, H.v.H., 1990. Effect of exchangeable potassium on the hydraulic conductivity and infiltration rate of some South African soils. Soil Sci. 149, 69–77. Magistad, O.C., 1945. Plant growth relations on saline and alkali soils. Bot. Rev. 11, 181–230.




Marchuk, A.G., Rengasamy, P., 2010. Cation ratio of soil structural stability (CROSS). In: 19th World Congress of Soil Science, Soil Solutions for a Changing World, 1–6 August, Brisbane, Australia, pp. 9–11. Marchuk, A., Rengasamy, P., 2011. Clay behaviour in suspension is related to the iconicity of clay-cation bonds. Appl. Clay Sci. 53, 754–759. Marchuk, A., Rengasamy, P., McNeill, A., Kumar, A., 2012. Nature of the clay-cation bond affects soil structure as verified by X-ray computed tomography. Soil Res. 50, 638–644. Oster, J.D., Jayawardane, N.S., 1998. Agricultural management of sodic soils. In: Sumner, M.E., Naidu, R. (Eds.), Sodic Soil: Distribution, Properties, Management and Environmental Consequences. Oxford University Press, New York, pp. 125–147. Oster, J.D., Sposito, G., 1980. The Gapon coefficient and the exchangeable sodium percentage-sodium adsorption ratio relation. Soil Sci. Soc. Am. J. 44, 258–260. Quirk, J.P., 2001. The significance of the threshold and turbidity concentrations in relation to sodicity and microstructure. Aust. J. Soil Res. 39, 1185–1217. Quirk, J.P., Schofield, R.K., 1955. The effect of electrolyte concentration on soil permeability. J. Soil Sci. 6, 163–178. Reeve, R.C., Bower, C.A., Brooks, R.H., Gschwend, F.B., 1954. A comparison of the effects of exchangeable sodium and potassium upon the physical condition of soils. Soil Sci. Soc. Proc. 18, 130–132. Rengasamy, P., Sumner, M.E., 1998. Processes involved in sodic behaviour. In: Sumner, M.E., Naidu, R. (Eds.), Sodic Soil: Distribution, Properties, Management and Environmental Consequences. Oxford University Press, New York, pp. 35–50. Rengasamy, P., Greene, R.S.B., Ford, G.W., Mehanni, A.H., 1984. Identification of dispersive behaviour and management of red-brown earths. Aust. J. Soil Res. 22, 413–431.

Rengasamy, P., Marchuk, A., 2011. Cation ratio of soils structural stability (CROSS). Soil Res. 49, 280–285. Rhoades, J.D., Kandiah, A., Mashali, A.M., 1992. The use of saline waters for crop production. FAO Irrig. Drain. Pap. 48. FAO, Rome, Italy. Rosenbrock, H.H., 1960. An automatic method for finding the greatest or least value of a function. Comput. J. 3, 175–184. Shainberg, L., Shalhevet, J. (Eds.), 1984. Soil Salinity Under Irrigation: Processes and Management. Springer-Verlag, Berlin, Germany. Smiles, D.E., Smith, C.J., 2004a. A survey of the cation content of piggery effluents and some consequences of their use to irrigate soils. Aust. J. Soil Res. 42, 231–246. Smiles, D.E., Smith, C.J., 2004b. Absorption of artificial piggery effluents by soil: a laboratory study. Aust. J. Soil Res. 42, 961–975. Sposito, G., 2008. The Chemistry of Soils, 2nd ed. Oxford University Press, New York. Sumner, M.E., 1993. Sodic soils: new perspectives. Aust. J. Soil Res. 31, 683–750. Sumner, M.E., Naidu, R. (Eds.), 1998. Sodic Soil: Distribution, Properties, Management and Environmental Consequences. Oxford University Press, New York. Tanji, K.K., Kielen, N.C., 2002. Agricultural drainage water management in arid and semi-arid areas. FAO Irrig. Drain. Pap. 61. FAO, Rome, Italy. Tyagi, N.K., Minhas, P.S., 1998. Agricultural Salinity Management in India. Director, Central Soil Salinity Research Institute, Karnal, India. U.S. Salinity Laboratory Staff, 1954. Diagnosis and Improvement of Saline and Alkali Soils. Handbook No. 60. US Department of Agriculture, Washington, DC. Wallender, W.W., Tanji, K.K., 2012. Agricultural Salinity Assessment and Management. ASCE Manuals and Reports on Engineering Practice No. 71. ASCE, Reston, VA. Vyshpolsky, F., Mukhamedjanov, K., Bekbaev, U., Ibatullin, S., Yuldashev, T., Noble, A.D., Mirzabaev, A., Aw-Hassan, A., Qadir, M., 2010. Agric. Water Manage 97, 1277–1286.
