Effects of sodium-contaminated wastewater on soil permeability of two New Zealand soils.
There is increasing anecdotal evidence from some land treatment sites in New Zealand that irrigating sodium-contaminated wastewaters onto soils may be causing soil structural problems and reduced permeability. In this study, the effect of irrigating such waste (derived from agricultural industries) on soil physical and chemical properties was investigated in an Allophanic Soil (Te Puninga silt loam) and a Gley Soil (Waitoa silt loam). Wastewater irrigation at the sites investigated had taken place for the previous 5 years, with sodium adsorption ratios (SAR) of the wastewater varying between 17 and 51 [(mmol/L).sub.0.5]. Increases in exchangeable sodium percentage (ESP) were recorded to 300 mm depth in both soils. At the soil surface (0-20 mm), ESP had increased to 31%, compared with 0.4% at control sites.
In laboratory studies using soil from the 0-20 mm layer in non-irrigated sites, leaching distilled water through repacked columns of the soil pretreated with various SAR solutions caused saturated hydraulic conductivity ([K.sub.sat]) to decrease below 100% at SAR greater than 3.5 and 8.5 for the Waitoa silt loam and Te Puninga silt loam, respectively. The decreases in [K.sub.sat] coincided with an increase in dissolved organic carbon (DOC) in collected leachate samples, and no dispersed clay was observed in the leachate. The laboratory studies would predict that effects of past irrigation of industrial wastewater at the study site would be measurable in the field due to the large ESPs that were recorded.
Saturated and unsaturated hydraulic conductivity measurements carried out at irrigated sites in the field showed no evidence of reduced conductivity in the surface soil until a pressure head of -120 mm was applied, the decrease being greater for the Te Puninga soil than the Waitoa. These results, along with the laboratory studies, suggest that whereas there may have been some structural deterioration in the soil matrix as a result of irrigation with the wastewater, macropore flow at higher moisture contents in the field was sufficient to overcome any adverse effects. It is suggested that laboratory studies using repacked soil may have limited use in predicting effects of Na-contaminated wastewater on soil hydraulic properties in structured soils. The results also further support suggestions that organic matter dissolution in Na-affected soils may affect soil physical properties.
Additional keywords: sodium, organic matter dissolution, soil structure, water movement.
The application of industrial liquid wastes to soil is becoming a popular alternative to direct disposal of wastes into surface waters. Many industrial wastewaters contain considerable sodium (Na) due to its use in both processing and cleaning. Tannery (Mason 1981), meat processing (Keeley and Quin 1979; Cooper and Russell 1992; Russell and Cooper 1992), fellmongery (Keeley and Quin 1979), and dairy (Barnett and Upchurch 1992) industrial wastes in New Zealand have all been reported to contain appreciable amounts of Na, with a sodium adsorption ratio (SAR; Sumner 1993) in the range 4-50 [(mmol/L).sup.0-5].
The effects of irrigating fresh water onto sodic soils, or saline irrigation water onto soils, are well documented in the literature (Shainberg et al. 1981; Sumner 1993). Most studies appear to have been conducted on soils from arid or semi-arid regions that have 2:1 clay minerals as an important component of the clay mineral fraction, low organic matter contents, and often poor natural structure. Many studies have shown that the soil exchangeable sodium percentage (ESP; Stunner 1993) generally increases linearly with SAR of irrigation water (Bower 1959; Harron et al. 1983; Curtin et al. 1995), often in accordance with Eqn 1 (Richards 1954):
(1) ESP = [100(-0.0126+01475 SAR)]/[1+(-0.0126+0.01475 SAR)]
An increase in soil ESP can, under certain conditions, cause soil structural deterioration through clay swelling and dispersion, and by the slaking of silt-sized micro-aggregates (Abu-Sharar et al. 1987; Curtin et al. 1994a; Amezketa and Aragues 1995).
The critical point at which clay dispersion occurs depends on the interaction of the soil exchangeable Na fraction (i.e. ESP) and the total electrolyte concentration (TEC) of the soil solution (Quirk and Schofield 1955). Clay dispersion can be a problem across a range of ESP values provided the TEC is below the critical flocculation concentration. Under these conditions, the thickness of the diffuse double layer increases, and the attractive forces between clay particles decrease, leading to dispersion. The blocking of water-conducting soil pores with dispersed clay particles can lead to reduced water and air movement, and root penetration.
The reduction in permeability in sodic soils by clay dispersion has been well documented, and can result from a decrease in surface infiltration through surface crusting (McIntyre 1957; Agassi et al. 1981; Kazman et al. 1983) or by a decrease in hydraulic conductivity of the soil through pore blockage (McNeal and Coleman 1966; Shainberg and Caiserman 1971; Pupisky and Shainberg 1979; Curtin et al. 1994a). Laboratory studies, using disturbed soil samples, have often shown that both saturated (Shainberg and Caiserman 1971; Pupisky and Shainberg 1979; Curtin et al. 1994a; Crescimanno et al. 1995) and unsaturated (Gardner et al. 1959; Malik et al. 1992; Crescimanno et al. 1995) hydraulic conductivity can decrease as a result of sodium accumulation in soils. Although several countries use critical ESP concentrations of up to 15 to define Na-affected soils, no universal critical conditions can be defined for structural deterioration in soils, with factors such as clay mineralogy (McNeal and Coleman 1966; Velasco-Molina et al. 1971; Frenkel et al. 1978; Curtin et al. 1994b), clay content (McNeal et al. 1968; Curtin et al. 1994b), sesquioxides and organic matter (Goldberg and Glaubig 1987), and soil pH (Frenkel et al. 1978; Suarez et al. 1984; Chiang et al. 1987; Miller et al. 1990) all affecting the response to Na accumulation. Recently, Crescimanno et al. (1995) suggested that a continuum may exist between soil structural properties and ESP, with an ESP as small as 2-5 causing adverse effects if low electrolyte concentrations are present in the soil solution.
In New Zealand, land treatment of liquid wastes is popular on farmland surrounding industrial sites. Liquid wastes differ from saline irrigation waters because they tend to contain a wider variety of cations and anions. Whilst theory would suggest that Nacontaminated liquid wastes are unlikely to cause adverse effects (due to high electrolyte concentrations when applied), there is concern in New Zealand that soil permeability has been adversely affected at some sites, especially during high rainfall periods in winter months, perhaps due to the long-term accumulation of Na in soils. We investigated the effects of irrigating Na-contaminated, industrial wastewater onto non-sodic soils derived from volcanic parent materials. The soils are different from most past studies because they have 1:1 clay minerals dominating the clay fraction and high organic matter contents. The liquid waste was derived from a mixture of agriculturally based industries, and records showed that SAR varied between 17 and 57 over the 5 years of irrigation prior to the study.
Most studies on the effects of Na in irrigation water have been conducted on repacked laboratory columns. In this study, a laboratory experiment was initially conducted to establish whether adverse effects of Na might be expected in soils that have different mineralogical composition and larger organic matter contents than have been studied in the past. A field study then investigated the effects of previous irrigation of Na-contaminated wastewater in situ, to establish whether studies in repacked laboratory columns could predict effects on soil containing their natural soil structure.
Materials and methods
Soils and site description
The study was carried out on a dairy farm 60 km north-east of Hamilton, New Zealand. The site, which is owned by an adjacent industrial company, consists of a rendering plant, hide and skin processing plants, and a meatworks. Liquid wastes from the industries are combined in holding ponds and regularly sprayirrigated onto the surrounding soils during summer. Direct discharge into local rivers is permitted during winter. Irrigation onto land occurs on a rotational basis (with 7.5 mm of wastewater applied in any 12-h irrigation period) dependent on the grazing regime of the dairy herd. Irrigation records showed that different areas of the farm had received different amounts of wastewater (up to a maximum of 300 mm) in the previous 5 years, and these records were used as a guide for initial sampling.
Two soil types exist in the irrigated areas: Te Puninga silt loam (Aquic Hapludand, Soil Survey Staff 1994; Allophanic Soil, Hewitt 1993), and Waitoa silt loam (Aquandic Haplaquept, Soil Survey Staff 1994; Gley Soil, Hewitt 1993). The soils are part of a toposequence formed on the low-lying ridge and swale topography of a relict floodplain. The parent material of both soils is volcanogenic alluvium and composite airfall tephra. The Waitoa silt loam is located in the swales and has impeded drainage, and a topsoil with a pH of 5.5 and organic carbon of 9.8%, and a clay fraction (27%) that comprises mainly halloysite and kaolinite. The Te Puninga silt loam was formed on the ridges and is better drained than the Waitoa soil, with a topsoil pH of 5.7, organic carbon content of 11.3%, and a clay content of 21%, which is predominantly allophane and kaolinite. Both soils have low topsoil bulk densities (0.79 and 0.84 Mg/[m.sup.3] for the Te Puninga and Waitoa soils, respectively), typical of soils derived from volcanic parent materials.
A preliminary site investigation was undertaken to determine soil chemical properties with depth on areas of the farm that had received different amounts of wastewater irrigation. Changes in soil chemical properties as a result of effluent irrigation were recorded up to 300 mm depth, with increases in exchangeable Na, EC, and ESP (Table 1). The largest increases in exchangeable Na were recorded at the surface. Other soil properties were generally unaffected. At the soil surface, the ESP ranged from [is less than] 0.4% in unirrigated areas to [is greater than] 30% in irrigated areas. Similar increases in ESP were recorded at depth, due, in part, to the relatively low CEC present, which enabled small amounts of exchangeable Na to have relatively large effects on ESP. The effects of effluent irrigation were generally greater in the Waitoa soil than in the Te Puninga soil. All further soil sampling and laboratory analyses reported in this paper were conducted on soil taken from the 0-20 mm depth.
Table 1. Effect of irrigation on soil chemical properties in the Te Puninga and Waitoa silt loams at three selected depths Values are means, with ranges in parentheses Soil Na(A) Ca(A) Mg(A) K(A) ([cmol.sub. c]/kg) 0-20 mm depth Te Puninga Non- 0.08 15.80 2.08 0.62 irrigated (0.056-0.11) (14.8-16.8) (1.9-2.2) (0.5-0.7) Irrigated 1.26 15.72 2.41 0.81 (0.4-3.05) (10-20) (1.7-3.0) (0.2-2.3) Waitoa Non- <0.01 12.50 1.40 1.68 irrigated -- (10-15) (1.2-1.6) (1.05-2.3 Irrigated 1.79 11.28 1.74 1.07 (0.57-3.87) (7-18.5) (1.2-2.4) (0.6-2.12) 100-120 mm depth Te Puninga Non- <0.01 7.1 0.93 0.02 irrigated -- (5.8-8.4) (0.92-0.94) (0-0.04) Irrigated 0.60 5.63 0.71 0.15 (0.01-1.22) (3.2-9.0) (0.4-1.58) (0.02-0.6) Waitoa Non- <0.01 4.75 0.63 0.58 irrigated -- (3.50-6.00) (0.40-0.85) (0.40-0.75) Irrigated 0.73 3.40 0.54 0.42 (0.38-1.53) (2.00-6.05) (0.30-0.85) (0.05-0.75) 300-320 mm depth Te Puninga Non- <0.01 2.1 0.27 0.07 irrigated -- (2-2.2) (0.2-0.34) (0-0.14) Irrigated 0.39 1.7 0.17 0.11 (0.17-0.57) (0.6-2.2) (0.04-0.3) (0-0.34) Waitoa Non- <0.01 1.5 0.68 0.35 irrigated -- (1.0-2.0) (0.35-1.0) (0.2-0.35) Irrigated 0.32 1.33 1.33 0.28 (0.02-0.78) (0.5-3.0) (0.5-3.0) (0-0.5) Soil CEC(A) EC(B) ESP pH(B) (dS/m) (%) 0-20 mm depth Te Puninga Non- 21 0.28 0.4 5.5 irrigated (20.6-21) (0.27-0.30) (0.3-0.5) (5.3-5.7) Irrigated 21 0.45 5.8 5.7 (17-27) (0.21-1.2) (2-16) (4.8-6.1) Waitoa Non- 17 0.46 0 5.6 irrigated (15-19) (0.34-0.57) -- (5.4-5.7) Irrigated 16 0.71 11.7 5.4 (12.4-23) (0.33-1.3) (4-31) (4.7-5.8) 100-120 mm depth Te Puninga Non- 9 0.13 0 5.5 irrigated (9.7-12.2) (0.84-0.18) -- (5.3-5.6) Irrigated 11 0.21 6.8 5.3 (5.5-13.5) (0.84-0.40) (1.7-15) (4.9-5.7) Waitoa Non- 8 0.25 0 4.8 irrigated (6.1-9.7) (0.164).33) -- (4.3-5.3) Irrigated 6 0.33 11.9 4.9 (4.1-9.0) (0.14-1.02) (2-22.3) (4.4-5.3) 300-320 mm depth Te Puninga Non- 4 0.082 0 5.7 irrigated -- (0.07-.095) -- -- Irrigated 4 0.106 11.2 5.6 (2.1-11.6) (0.069-.188) (2-19.5) (5.2-6.0) Waitoa Non- 2 0.077 0 4.5 irrigated (0.9-4.1) (0.058-.096) -- (4.4-4.6) Irrigated 3 0.157 13.6 4.9 (0.9-7.7) (0.058-249) (1-36.5) (4.4-5.6) (A) Unbuffered 0.01 M silver thiourea (Blakemore et al. 1987). (B) Soil: water 1:5.
Saturated hydraulic conductivity was measured using sieved ([is less than] 3.75 mm) field-moist soil, collected from 0-20 mm depth at non-irrigated sites, repacked into polyvinyl chloride cores (65 mm internal diameter by 120 mm length). A fine stainless steel mesh was glued to the base and overlain by filter paper to prevent loss of soil. The soil was packed uniformly into the cores in 10-mm increments to a depth of 40 mm, to a dry bulk density similar to that existing in the field. Cores were slowly saturated overnight through capillary action, by standing them in distilled water. Filter paper was also placed on the soil surface to prevent excessive disturbance by the infiltrating solution.
Initially, all cores were leached with distilled water (0.07 dS/m) under a constant head of 40 mm until a steady state flux was observed, to establish the baseline initial steady state hydraulic conductivity ([K.sub.i]) for each core. Solutions of varying SAR (Table 2) were then passed through the columns to displace the resident water, and the columns were saturated from the base overnight in the SAR solution to ensure saturation of exchange sites with Na. The next morning, the soil in the columns was leached with SAR solution, under a constant head of 40 mm, until a steady state flux was again attained (approximately 3.0 pore volumes of drainage), at which point the saturated hydraulic conductivity ([K.sub.s]) was calculated. This amount of drainage corresponded to a time period of approximately 60 min for most of the treatments. Distilled water was then reintroduced to the columns and leached until the flux of solution from the columns appeared to become steady (approximately 8-12 pore volumes of drainage), so that the final hydraulic conductivity ([K.sub.f]) could be calculated. The hydraulic conductivity at the different time periods was calculated by recording the flux density in 4 replicate cores for each treatment and applying Darcy's Law. To allow for differences in hydraulic conductivity between individual cores, the relative saturated hydraulic conductivity obtained for SAR solutions or the final distilled water treatment was calculated as a percentage of [K.sub.i]:
(2) Relative K = [K.sub.s,f]/[K.sub.i] x 100(%)
The SAR solutions used were representative of the range of SAR levels of the liquid waste irrigated in the field. The solutions were prepared by maintaining a constant Na concentration (0.32 g/L), which is typical of liquid waste irrigated in the field at the site, and varying the amounts of Ca and Mg. A decrease in SAR is also reflected in an increase in electrolyte concentration, which is similar to what would occur if industry was required to decrease the SAR for land disposal of its liquid waste (as it would be a more feasible option for industry to increase divalent cation concentrations in effluent rather than decrease Na concentrations). The ESP of soil following the application of different SAR solutions to the soil (prior to the reintroduction of distilled water) was not measured. However, Lieffering and McLay (1996) recently showed that when similar soils were equilibrated with solutions of varying SAR, the resulting ESP of the soil could be accurately predicted using Eqn 1 given above. Properties of the infiltrating solutions and resulting predicted soil ESP are given in Table 2.
Table 2. Properties of infiltrating solutions and soil ESP SAR Electrical Total Predicted [(mmol/L).sup.0.5] conductivity electrolyte ESP(A) (dS/m) concentration (%) ([mmol.sub.c]/L) 0 0.07 0 0 1.5 7.7 157 1 4 3.1 37 4 8 1.9 19 10 16 1.6 15 18 40 1.5 14 37 80 1.5 14 54 (A) Predicted from SAR of infiltrating solutions using Eqn 1.
Discoloration of leachate during experiments suggested organic matter dissolution. Leachate samples were, therefore, retained for analysis from different treatments. Total carbon and dissolved organic carbon (= total carbon [is less than] 0.2 mm) were measured by a Shimadzu, TOC 5000 automated total organic carbon analyser.
The curves of relative hydraulic conductivity v. time conformed in shape to a reverse sigmoid, described by Eqn 3:
(3) K/[K.sub.i] = K/[K.sub.i(min)] + a/[1+exp.sup.-(x-x0/b)]
where K/[K.sup.i] is relative hydraulic conductivity at given time, t, relative to the initial ([K.sub.i]); a is the absolute increase in K/[K.sub.i] above K/[K.sub.i(min)] following introduction of the SAR solution (i.e. K/[K.sub.i(max)] = K/[K.sub.i(min)] + a); K/[K.sub.i(min)] is the final steady state hydraulic conductivity; b represents the transition between x values at 75% and 25% of the distance from K/[K.sub.i(min)] to the maximum of smoothed data (i.e. the linear portion of the curve); and x0 is the amount of drainage when 50% of the decrease in K/[K.sub.i] occurs. Curve-fitting parameters were computed for each replicate soil core by using Eqn 2. Treatment means were then compared using analysis of variance to determine whether curve-fitting parameters were different between treatments.
Saturated hydraulic conductivity ([K.sub.sat] was measured during winter using a modified twin ring method (Scotter et al. 1982), at 22 locations that the preliminary study had shown would provide a range in ESP. In this method, pairs of rings with different diameters are often used. In the larger ring, the infiltration rate is predominantly driven by the [K.sub.sat], whereas in the smaller ring, the infiltration rate is dominated by sorptivity (S). In our study, S was negligible due to the high moisture contents when measurements were made. Hence, only one large ring (175 mm diameter) was necessary to measure steady-state flux. The rings were gently pressed into the soil surface and ponded with water to a depth of 20 mm. Any leaks around the base of the rings were sealed using bentonite on the outside edge. Water levels were maintained to a precise datum point and changes in volume with time were recorded. Measurements were taken for up to 2 h after steady state had been achieved (usually within 1 h after commencing infiltration). Five replicate measurements were made at each site.
Unsatururated hydraulic conductivity
Unsaturated hydraulic conductivity ([K.sub.unsat]) was measured at 7 locations with disc permeameters by using the multiple head method of Ankeny et al. (1991). The method is based on Wooding's solution for infiltration from a circular source with a constant pressure head at the soil surface (Wooding 1968). If K is assumed to have an exponential function, then:
(4) [K.sub.1,2] = [K.sub.s] exp([[Alpha]h.sub.1,2])
where [K.sub.1] and [K.sub.2] are the hydraulic conductivities at pressure heads of [h.sub.1] and [h.sub.2], respectively; [K.sub.s] is the saturated hydraulic conductivity; and ct is the exponential slope. The steady state volumetric infiltration rate Q ([mm.sup.3]/h) for a given pressure head is given by:
(5) Q = K([Pi][r.sup.2]+4r/[Alpha])
where r is the disk radius. A measure of the soil's capillarity from the 2 flow observations at each pressure head can be estimated from:
(6) [Alpha] = 1n([Q.sub.1]/[Q.sub.2])/ [h.sub.1] - [h.sub.2]
where [Q.sub.1] and [Q.sub.2] are the volumetric infiltration rates at pressure heads of [h.sub.1] and [h.sub.2], respectively. Rearrangement of Eqn 5 enables K to be determined at the different pressure heads. In this experiment, pressure heads of-120, -40, and -10 mm were used, giving -120 and -40 mm, and -40 and -10 mm pairs.
A `characteristic' mean pore size (lm) for soil with different ESP was calculated using Eqn 7 (Clothier et al. 1995):
(7) lm = [Sigma][Alpha]/pg
where [Sigma] is the surface tension of water, P is the density of water, and g is the acceleration due to gravity.
[K.sub.unsat] was measured at 7 field locations that had different ESP values, on both the Waitoa and Te Puninga soils (see Table 3). Tension disc permeameters were placed on a thin layer of sand (to facilitate contact) upon the soil surface. Readings were initially made with the smallest pressure head. Triplicate [K.sub.unsat] measurements were made at each site.
Table 3. Mean `characteristic' pore size for Waitoa silt loam and Te Puninga silt loam Values represent means of triplicate measurements in each of 7 paddocks. Within each soil type, values followed by the same letter are not significantly different at P = 0.05 Te Puninga silt loam ESP lm lm (%) (<-40 mm) (>-40 mm) 0 0.09a 0.64a 3.2 0.20b 1.53b 6.1 0.19ab 1.41ab 6.2 0.23b 1.71b 7.2 0.15ab 1.15ab 7.3 0.19b 1.45b 16.4 0.18ab 1.35ab Waitoa silt loam ESP lm lm (%) (<-40 mm) (>-40 mm) 0 1.07a 1.07a 3.8 0.27b 0.42b 9.5 0.20bc 0.59b 11.1 0.15cd 0.88a 14.8 0.15cd 1.05a 21.9 0.21c 0.56b 30.8 0.10ad 0.57b
Effects of Na on repacked soil
For both the Te Puninga and Waitoa soils, K/[K.sub.i] remained close to 100% in cores that received only distilled water (SAR = 0), even after 6 pore volumes of drainage. The addition of SAR solutions in the range 1.5-80 enhanced K/[K.sub.i] of both soils (Fig. 1). However, the reintroduction of distilled water, following leaching with SAR solutions, caused a marked reduction in K/[K.sub.i], the magnitude of which depended on the SAR treatment. In the Te Puninga soil, maximum enhancement in hydraulic conductivity during initial leaching with SAR solutions, and least decrease with the reintroduction of distilled water, occurred with SAR of 1.5. Intermediate enhancement (with SAR solutions) and decrease (distilled water) occurred with SAR solutions 4 and 8, and least enhancement and maximum decrease with solutions of SAR [is greater than or equal to] 16. A similar result occurred for the Waitoa soil, except the maximum decrease with the re-introduction of distilled water occurred for solution SAR [is greater than or equal to] 8.
An analysis of the curve-fitting parameters showed further differences between treatments. The Te Puninga silt loam had a larger overall increase in K/[K.sub.i(max)] than the Waitoa silt loam (P [is less than] 0.05). In both soil types, the transition period (b) required to reach K/[K.sub.i(min)] after distilled water was introduced was generally shorter for high SAR solutions than for low SAR solutions. The exception was with an SAR of 1.5 in the Te Puninga silt loam, which caused K/[K.sub.i(min)] to be reached more rapidly than with solutions of SAR 4 and 8, but was similar to solutions with SAR of 16, 40, and 80. In general, the Te Puninga silt loam reached K/[K.sub.i(min)] more quickly than the Waitoa silt loam (P [is less than] 0.05), although the Waitoa soil had a greater overall decrease in K/[K.sub.i(min)] than the Te Puninga: soil (P [is less than] 0.05).
The relationship between K/[K.sub.i(min)] and SAR is shown in Fig. 2. The SAR at which K/[K.sub.i](min) = 100% (i.e. was the same as the initial relative hydraulic conductivity) was approximately 3.5 and 8.5 for the Waitoa silt loam and Te Puninga silt loam, respectively.
Leachate solutions were clear when repacked soil cores were leached with SAR solutions, but turned yellow when leached with distilled water. No suspended material was visible, suggesting that if clay dispersion occurred, it was not leached from the cores. Total carbon in the leachate increased with pretreatment SAR up to 8 in the Te Puninga soil, and 16 in the Waitoa soil. At larger SARs, total carbon remained constant in the leachate, although it was approximately 30% greater in the Waitoa soil than in the Te Puninga soil. A 1:1 relationship existed between total carbon and dissolved organic carbon in the leachate, confirming that the carbon was mainly in a dissolved form (Fig. 3).
In situ saturated hydraulic conductivity
A large range of [K.sub.sat] was observed in different paddocks, with no relationship evident between [K.sub.sat] and ESP in either soil type. Investigations of other related Na indices in the soil (e.g. exchangeable Na; exchangeable Na:exchangeable bases; exchangeable Na:electrical conductivity) did not improve the relationship.
In situ unsaturated hydraulic conductivity
Unsaturated hydraulic conductivity, [K.sub.unsat], at -120 mm ([K-.sub.120mm]) was found to be significantly correlated (P [is less than] 0.01) with ESP in both soil types (Fig. 4). Exchangeable Na, and the ratio exchangeable Na:exchangeable bases also predicted (P [is less than] 0.05) [K-.sub.120mm] with a similar relationship to that shown with ESP. However, other indices tested (e.g. exchangeable Na:electrical conductivity) resulted in poorer prediction, and no relationship between any soil Na index and hydraulic conductivity was evident at the larger potentials.
Two-domain model of near saturated hydraulic conductivity
The 2-domain model used to fit hydraulic conductivity data measured across 3 pressure heads suggests that the functioning pores are quite different in the -10 to -40 mm than the -40 mm to -120 mm suction range in the soils (Fig. 5). The large decrease in [K.sub.unsat] between -10 and-40 mm heads indicates the importance of macropore flow during or near saturated flow conditions.
In both the Te Puninga and Waitoa soils, ESP appeared to influence the hydraulic character of the macro-mesopore flow domains only at the largest ESPs, with no clear trends in either the macropore or mesopore flow domains at lower ESPs (Fig. 5). The `characteristic' mean pore size (lm; Eqn 6) was calculated for the macropore and mesopore flow domains at each ESP and soil type (Table 3). Although differences in lm existed in both the Waitoa silt loam and the Te Puninga silt loam (P [is less than] 0.05), no clear trend with ESP was evident.
The irrigation of liquid wastes in the field over 4 years resulted in a large increase in exchangeable Na and ESP in some areas. Increases in exchangeable Na were observed to at least 300 mm depth. The increases in exchangeable Na, however, do not appear to have affected other resident cations in a major way, and indicate that Na must compete poorly for exchange sites. This is consistent with observations we have recently made in other land treatment systems in New Zealand (e.g. Hopkins 1997). The larger increases in ESP that were recorded in the Waitoa silt loam than in the Te Puninga silt loam may be related to mineralogical differences, although it is possible the topographic setting may also have influenced the results. The Waitoa silt loam is formed in lower lying areas than the Te Puninga silt loam and any overland or subsurface flow of irrigated liquid waste would probably accumulate in this soil.
The studies on repacked soil indicated that liquid wastes containing Na can potentially cause problems for water movement in the soils. The [K.sub.sat] of Waitoa and Te Puninga soils decreased below 100% by SAR levels in excess of 3.5 and 8.5, respectively (Fig. 2). However, effects on soil hydraulic properties were only apparent in the field under the most unsaturated flow conditions studied ([K.sub.-120mm]). The lack of readily observable effects of Na in the field may be attributed to either strongly structured soils that are not readily degraded, or variability and macropore effects masking any soil matrix effects which occurred.
The laboratory study used the upper 20 mm of soil, repacked into 40-mm-deep cores. In the field situation this layer would not have been so uniform, with a larger proportion of macropores occurring either naturally or induced through, for example, earthworm activity or cracks and surface disturbances caused by animal activity. Indeed, the amount of variability in water movement that was observed in the field soils was substantially higher than in repacked soils, especially at moisture conditions close to saturation (coefficients of variation in field soils were up to 20 times higher than in repacked soil). It has been reported that the presence of a single macropore in field soils can transmit a large proportion of soil water at moisture contents close to saturation (Beven and Germann 1982) and mask effects that have occurred in the soil matrix. This may well explain why effects were only apparent in the field situation at [K.sub.-120mm], where the influence of macropores was less and hydraulic conductivity was controlled more by the characteristics of the soil matrix.
The 2-domain model of [K.sub.unsat] showed no clear relationship with ESP (Fig. 5), although the results do indicate that the functioning pores between 3.0 and 0.75 mm in diameter (i.e. pores capable of flowing in the -10 to -40 mm potential range) are quite different from those in the 0.75-0.25 mm range (i.e. pores capable of flowing in the -40 to -120 mm potential range). Therefore, between -10 and -120 mm pressure potential, there is a break in the size of functioning pores. Messing and Jarvis (1993) found likewise, with their breakpoints occurring between -45 and -65 mm pressure potential. The break-point pressure head (Messing and Jarvis 1993) separating macropore from mesopore flow domains, experimentally defined in this study as -40 mm, may be better resolved by measurement of [K.sub.unsat] at a range of pressure heads either side of-40 mm. Several authors (e.g. Clothier and Smettem 1990; Sauer et al. 1990; Jarvis and Messing 1995) have shown that the characteristic mean pore size (lm) for a soil may be indicative of changes to soil structural properties. However, no relationship between lm and ESP was established in either the Te Puninga or Waitoa soils. The 2-line exponential model is constructed in piecewise segments (Clothier and Smettem 1990), and the use of pressure heads near saturation may, therefore, cause any effects of Na on the soil matrix to be masked due to macropore flow.
In the laboratory columns, the initial increase of K/[K.sub.i] following application of the SAR solutions in both soils is consistent with diffuse double-layer theory, which predicts that an increasing solution electrolyte concentration will result in compression of the diffuse double-layer, encouraging further flocculation and less swelling and resulting in increased soil permeability. After the decrease of K/[k.sub.i]following the introduction of distilled water, final K/[k.sub.i] (i.e. K/[K.sub.i(min)]) appeared to become constant at SARs greater than 16 and 23 for the Waitoa silt loam and the Te Puninga silt loam, respectively (Fig. 2). This upper limit of SAR was represented by a maximum decrease in K/[K.sub.i(min)] of approximately 20% (Waitoa soil) and 15% (Te Puninga soil). In contrast, the maximum decrease in the field was 41% and 35% in the Waitoa silt loam and the Te Puninga silt loam, respectively. Whereas a similar decrease in relative hydraulic conductivity was recorded in laboratory and field measurements for the Waitoa soil (Table 4), different decreases occurred in the Te Puninga soil and indicate that disturbed soil columns may have limited applicability for measuring effects of Na contamination in some soils.
Table 4. Effect of ESP on relative hydraulic conductivity for the Te Puninga silt loam and the Waitoa silt loam Decrease in K (%) ESP(%) Waitoa silt loam Te Puninga silt loam Laboratory(A) Field(B) Laboratory(A) Field(B) 0 4 0.1 10 0.4 5 5 3 14 2 10 6 6 19 3 15 9 8 23 4 20 >16 12 >23 6 (A) ESP predicted from SAR values using Richards's equation (Richards 1954). (B) ESP read from Fig. 4.
We suggest that the decreases in hydraulic conductivity following re-introduction of distilled water can be attributed to the effects of Na, and not the electrical conductivity (EC) of the solutions. Distilled water passing through the columns would decrease the EC of the solution in interstitial pore spaces. The addition of SAR 0 solution (EC~0) did not lead to a decrease in K/[k.sub.i] below 100% as the amount of drainage increased. However, the addition of SAR solutions greater than 8.0 (Waitoa soil) and 16.0 (Te Puninga soil) did cause K/[K.sub.i] to decrease below 100% at larger volumes of drainage. After 8-12 pore volumes of drainage, the EC of solution within the soil following the prior application of these SAR treatments may have approximated that of the SAR 0 treatment, but obviously it would not be possible for the EC to be have been less than occurred in the SAR 0 treatment. Also, we cannot postulate why a lower EC (i.e. at the higher SAR) would result in larger amounts of dissolved organic carbon being leached than was observed with the higher EC (i.e. lower SAR) solutions.
Typically, clay dispersion and swelling are implicated as causes of decreased hydraulic conductivity in Na-affected soils (Quirk and Schofield 1955; McNeal and Coleman 1966; Agassi et al. 1981; Cass and Sumner 1982; Curtin et al. 1994b). Recently, however, organic matter has been shown to have an important influence on clay dispersion in Australian soils, causing clay dispersion to be either enhanced or prevented depending on the type of organic matter and its selectivity for Na (Skene and Oades 1995; Nelson and Oades 1998; Nelson et al. 1999).
In this study, there was no evidence of dispersed clay in the leachate and it appears that organic matter dissolution may have been an important factor causing the collapse of aggregates and associated decrease in hydraulic conductivity. We suggest that even in non-sodic soils, sodium-contaminated wastewater may result in increased organic matter dissolution of surface soil at Na concentrations typical of industrial wastes, and that the amount of dissolution increases as the SAR of the solution increases, causing adverse structural changes within the soil matrix. The blocking of soil micropores may result in a greater incidence of macropore flow, causing faster leaching of surface-applied solutes (McLay et al. 1991). It has also been suggested that dissolution and leaching of organic ions may promote the transport of heavy metals and pesticide residues through soils (Skene and Oades 1995), thereby potentially contaminating ground and surface waters. The larger amount of organic carbon dissolved in the Waitoa silt loam compared with the Te Puninga silt loam possibly indicates that phyllosilicate clay minerals may be more susceptible to organic carbon dissolution in Na-affected soils than in soils dominated by allophanic clay minerals, supporting the suggestion that interactions between organic matter and clay are important to consider for predicting sodium effects in soils (Nelson et al. 1999).
Our results suggest that the irrigation of sodium-contaminated liquid wastes onto soils, followed by rainfall or fresh water irrigation (i.e. electrolyte dilution), may adversely affect hydraulic properties, particularly at or near the soil surface. In the past, clay swelling and dispersion have commonly been reported to cause soil structural deterioration in Na-affected soils. However, we suggest that in some soils derived from volcanic parent materials, at least, dissolution of organic matter may also lead to changes in the soil matrix adversely affecting soil physical properties. Where only a shallow surface layer of soil is affected, effects may be difficult to gauge in the field, where macropore flow, combined with disruption of the soil surface, will readily provide flow paths at high moisture contents to offset any adverse effects on soil structure. Unsaturated hydraulic conductivity at a pressure head of -120 mm was found to be a more sensitive index of structural deterioration than either saturated hydraulic conductivity, or unsaturated hydraulic conductivity at higher pressure potentials. Whereas there did not appear to be any key ESP levels at which any adverse effects could be expected, SAR levels in irrigated effluent of greater than 3.5 and 8.5 for the Waitoa and Te Puninga soils, respectively, caused hydraulic conductivity to decrease below 100% in laboratory studies.
We thank Mr Volker Turk who provided valuable assistance in the field and the Foundation of Research Science and Technology for providing scholarship funding for this study.
Abu-Sharar TM, Bingham FT, Rhoades JD (1987) Stability of soil aggregates as affected by electrolyte concentration and composition. Soil Science Society of America Journal 51, 309-314.
Agassi M, Morin J, Shainberg I (1981) Effect of electrolyte concentration and soil sodicity on infiltration rate and crust formation. Soil Science Society of America Journal 45, 848-851.
Amezketa E, Aragues R (1995) Hydraulic conductivity, dispersion and osmotic explosion in arid-zone soils leached with electrolyte solutions. Soil Science 159, 287-293.
Ankeny MD, Ahmed M, Kaspar TC, Horton R (1991) Simple field method for determining unsaturated hydraulic conductivity. Soil Science Society of America Journal 55, 467-470.
Barnett JW, Upchurch GC (1992) Irrigation of wastewater from the manufacturing dairy industry onto pasture. In `The use of wastes and byproducts as fertilizers and soil amendments for pastures and crops'. (Eds PEH Gregg, LD Currie) pp. 195-207. Occasional Report No. 6. (Fertilizer and Lime Research Centre, Massey University: New Zealand)
Beven K, Germann P (1982) Macropores and water flow in soils. Water Resources Research 18, 1311-1325.
Blakemore LC, Searle PL, Daley BK (1987) `Methods for chemical analysis of soils.' New Zealand Soil Bureau Scientific Report No. 80.
Bower CA (1959) Cation-exchange equilibria in soils affected by sodium salts. Soil Science 88, 32-35.
Cass A, Sumner ME (1982) Soil pore structural stability and irrigation water quality: 1. Empirical sodium stability model (1). Soil Science of America Journal 46, 503-507.
Chiang SC, Radcliffe DE, Miller WP, Newman KD (1987) Hydraulic conductivity of three southeastern soils as affected by sodium, electrolyte concentration, and pH. Soil Science Society of America Journal 51, 1293-1299.
Clothier BE, Heng L, Magesan GN, Vogeler I (1995) The measured mobile-water content of an unsaturated soil as a function of hydraulic regime. Australian Journal of Soil Research 33, 397-414.
Clothier BE, Smettem KRJ (1990) Combining laboratory and field measurements to define hydraulic properties of soil. Soil Science Society of America Journal 54, 299-304.
Cooper RN, Russell JM (1992) The New Zealand Meat Processing Industry, present effluent practices and future directions. In `Proceedings of the 1992 Food Industry Environmental Conference, Atlanta, Georgia'. pp. 95-108.
Crescimanno G, Iovino M, Provenzano G (1995) Influence of salinity and sodicity on soil structural and hydraulic characteristics. Soil Science Society of America Journal 59, 1701-1708.
Curtin D, Steppuhn H, Mernut AR, Selles F (1995) Sodicity in irrigated soils in Saskatchewan: chemistry and structural stability. Canadian Journal of Soil Science 75, 17-184.
Curtin D, Steppuhn H, Selles F (1994a) Clay dispersion in relation to sodicity, electrolyte concentration and mechanical effects. Soil Science Society of America Journal 58, 955-962.
Curtin D, Steppuhn H, Selles F (1994b) Structural stability of chernozemic soils as affected by exchangeable sodium and electrolyte concentration. Canadian Journal of Soil Science 74, 157-164.
Frenkel H, Goertzen JO, Rhoades JD (1978) Effects of clay type and content, exchangeable sodium percentage, and electrolyte concentration on clay dispersion and soil hydraulic conductivity. Soil Science Society of America Journal 42, 32-39.
Gardner WR, Mayhugh MS, Goertzen JO, Bower CA (1959) Effect of electrolyte concentration and exchangeable sodium percentage on diffusivity of water in soils. Soil Science 88, 270-273.
Goldberg S, Glaubig RA (1987) Effect of saturating cation, pH, and aluminium and iron oxides on the flocculation of kaolinite and montmorillonite. Clays and Clay Minerals 35, 220-227.
Harron WRA, Webster GR, Cairns RR (1983) Relationship between exchangeable sodium and sodium adsorption ratio in a Solonetzic soil association. Canadian Journal of Soil Science 63, 461-467.
Hewitt AE (1993). `New Zealand soil classification.' Landcare Science Series No. 1. (Manaaki-Wheua Press: Lincoln, NZ)
Hopkins K (1997) Chemistry of soil at the Rotorua Land Treatment System after four years of effluent irrigation. Unpublished MSc Thesis, University of Waikato.
Jarvis NJ, Messing I (1995) Near-saturated hydraulic conductivity in soils of contrasting texture measured by tension infiltrometers. Soil Science Society of America Journal 59, 27-34.
Kazman Z, Shainberg I, Gal M (1983) Effect of low levels of exchangeable sodium and applied phosphogypsum on the infiltration rate of various soils. Soil Science 135, 184-192.
Keeley GM, Quin BF (1979) The effects of irrigation with meatworks-fellmongery effluent on water quality in the unsaturated zone and shallow aquifer. Progress in Water Technology 11, 369-386.
Lieffering RE, McLay CDA (1996) Effects of high pH solutions with large monovalent cation concentrations on cation exchange properties. Australian Journal of Soil Research 34, 229-242.
Malik M, Mustafa MA, Letey J (1992) Effect of mixed Na/Ca solutions on swelling, dispersion and transient water flow in unsaturated montmorillonite soils. Geoderma 52, 17-28.
Mason I (1981) Tannery effluent. In `Proceedings of Water and Wastewater Treatment Operators Course'. Dunedin, New Zealand, 11 February. (NZ Leather and Shoe Research Association: Palmerston North, NZ).
McIntyre DS (1957) Permeability measurements of soil crusts formed by raindrop impact. Soil Science 85, 185-189.
McLay CDA, Cameron KC, McLaren RG (1991) Effect of time of application and continuity of rainfall on leaching of surface-applied nutrients. Australian Journal of Soil Research 29, 1-9.
McNeal BL, Coleman NT (1966) Effect of solution composition on soil hydraulic conductivity. Soil Science Society of America Proceedings 30, 308-312.
McNeal BL, Layfield DA, Norvell WA, Rhoades JD (1968) Factors influencing hydraulic conductivity of soils in the presence of mixed-salt solutions. Soil Science Society of America Proceedings 32, 187-190.
Messing I, Jarvis NJ (1993) Temporal variation in the hydraulic conductivity of a tilled clay soil as measured by tension infiltrometers. Journal of Soil Science 44, 11-24.
Miller WP, Frenkel H, Newman KD (1990) Flocculation concentration and sodium/calcium exchange of kaolinitic soil clays. Soil Science Society of America Journal 54, 346-351.
Nelson PN, Baldock JA, Clarke P, Oades JM, Churchman GJ (1999) Dispersed clay and organic matter in soil: their nature and associations. Australian Journal of Soil Research 37, 289-315.
Nelson PN, Oades JM (1998) Organic matter, sodicity and soil structure. In `Sodic soils. Distribution, properties, management, and environmental consequences'. (Eds ME Sumner, R Naidu) pp. 51-75. (Oxford: New York)
Pupisky H, Shainberg I (1979) Salt effects on the hydraulic conductivity of a sandy soil. Soil Science Society of America Journal 43, 429-433.
Quirk JP, Schofield RK (1955) The effect of electrolyte concentration on soil permeability. Journal of Soil Science 6, 163-178.
Richards LA (1954) `Diagnosis and improvement of saline and alkali soil.' USDA Handbook No. 60. US. (Government Printing Office: Washington)
Russell JM, Cooper RN (1992) The use of meat processing effluents for irrigation of pasture. In `The use of wastes and byproducts as fertilisers and soil amendments for pasture and crops'. (Eds PEH Gregg, LD Currie) pp. 208-213. Occasional Report No. 6. (Fertiliser and Lime Research Centre: Massey University, NZ)
Sauer TJ, Clothier BE, Daniel TC (1990) Surface measurements of the hydraulic properties of a tilled and untilled soil. Soil and Tillage Research 15, 359-369.
Scotter DR, Clothier BE, Harper ER (1982) Measuring saturated hydraulic conductivity and sorptivity using twin rings. Australian Journal of Soil Research 20, 295-304.
Shainberg I, Caiserman, A (1971) Studies on the Na/Ca montmorillonite systems 2. The hydraulic conductivity. Soil Science 111,276-281.
Shainberg I, Rhoades JD, Prather RJ (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.
Skene TM, Oades JM (1995) The effects of sodium adsorption ratio and electrolyte concentration on water quality: laboratory studies. Soil Science 159, 65-73.
Soil Survey Staff (1994) `Keys to soil taxonomy.' 6th Edn. United States of America, Department of Agriculture, Soil Conservation Service, Agricultural Handbook. (Pochahontas: Blacksburg, VA)
Suarez DL, Rhoades JD, Lavado R, Grieve CM (1984) Effect of pH on saturated hydraulic conductivity and soil dispersion. Soil Science Society of America Journal 48, 50-55.
Sumner ME (1993) Sodic soils: new perspectives. Australian Journal of Soil Research 31,683-750.
Velasco-Molina HA, Swoboda AR, Godfrey CL (1971) Dispersion of soils of different mineralogy in relation to sodium adsorption ratio and electrolyte concentration. Soil Science 111,282-287.
Wooding RA (1968) Steady state infiltration from a shallow circular pond. Water Resources Research 4, 1259-1273.
Manuscript received 13 July 1999, accepted 15 December 2000
J. C. Menneer(A), C. D. A. McLay(B), and R. Lee(C)
(A) AgResearch, Private Bag 3123, Hamilton, New Zealand; email: email@example.com
(B) Department of Earth Sciences, The University of Waikato, Private Bag 3105, Hamilton, New Zealand.
(C) Manaaki Whenua - Landcare Research, Private Bag 3127, Hamilton, New Zealand.
|Printer friendly Cite/link Email Feedback|
|Author:||Menneer, J. C.; McLay, C. D. A.; Lee, R.|
|Publication:||Australian Journal of Soil Research|
|Article Type:||Statistical Data Included|
|Date:||Jul 1, 2001|
|Previous Article:||Explaining groundwater hydrographs: separating atypical rainfall events from time trends.|
|Next Article:||Differences in sorption behaviour of carbaryl and phosalone in soils from Australia, Pakistan, and the United Kingdom.|